Our primary aim is to explore whether among-individual variation in T modulated cooperative behaviour is associated with differences in gene expression. We have 5 main traits that were measured.

  • Status: Territorial or Floater
  • Testosterone concentration
    • mean_T is best estimate of among individual variation in T across all birds (’r summary(behav$mean_T)`). This measure is repeatable (Ryder et al 2020) within individuals and could be considered a hormonal phenotype.
  • Strength
  • Mean number of social interactions per 10 hr day as detected by telemetry.

Here, I take the filtered dataset from the QC process, and run a whole brain analysis, as well as individual tissue analysis. I also explore the relationships between our explanatory variables.

key<- read.csv("../data_filtered/data_key_Parsed_ReplicatesRemoved.csv")
behav<- read.csv("../data_unfiltered/PIPFIL_T_and_behav_data.csv")
rownames(key)<- key$X

key$Color_ID<- sub("/","", key$Color_ID)
key<- plyr::rename(key, replace=c("Color_ID"="colorID"))
behav$status<- plyr::revalue(behav$status, c("floa"="floater", "terr"="territorial"))
key_behav<- merge(key, behav, by="colorID")
key_behav<- key_behav[!is.na(key_behav$last_behav),]


#create a data.frame with all of the observations. 
not_in_behav<- key[!key$X %in% key_behav$X,]
cols_not_key<- colnames(key_behav)[!colnames(key_behav) %in% colnames(key)]
cols_not_key_df<- data.frame(matrix(NA, nrow = nrow(not_in_behav), ncol = length(cols_not_key)))
colnames(cols_not_key_df)<- cols_not_key
not_in_behav<- cbind(not_in_behav, cols_not_key_df)

key_behav<- rbind(key_behav, not_in_behav)

rownames(key_behav)<- key_behav$X


key_behav$Tissue<- factor(key_behav$Tissue, levels=c("GON","PIT","VMH","AH","PVN","POM","ICO","GCT","AI","TNA","LS", "BSTm"))
key_behav<- key_behav[order(rownames(key_behav)),]
key_behav$Class<- as.factor(key_behav$Class)
key_behav$Class<- revalue(key_behav$Class, replace=c("SCB floater"="Predefinitive floater", "DCB floater "="Definitive floater", "DCB territorial"="Territorial"))
key_behav$Class<- factor(key_behav$Class, levels=c("Predefinitive floater", "Definitive floater", "Territorial"))
rownames(key_behav)<- key_behav$sampleID
key_behav<- key_behav[order(key_behav$sampleID),]
key_behav$Year<- as.factor(key_behav$Year)
key_behav$Batch<- as.factor(key_behav$Batch)
key_behav$Status<- as.factor(key_behav$Status)

key_behav_unique<- key_behav[!duplicated(key_behav$colorID),]

#read the raw count data
data<- read.csv("../data_filtered/data_RawCounts_all_ReplicatesRemoved_antisense_V2.csv")
data$X[data$X=="LOC113993669"] <- "CYP19A1"
data$X[data$X=="LOC113983511"] <- "OXT"
data$X[data$X=="LOC113983498"] <- "AVP"
data$X[data$X=="LOC113982601"] <- "AVPR2"
rownames(data)<- data$X
data$X<- NULL

### Gene Ontology


go_terms<- read.csv("../GO_annotations/pfil_GO_key_raw.csv")
go_terms<- plyr::rename(go_terms, replace=c("GeneID"="gene"))
go2gene_bp<- go_terms[which(go_terms$Aspect=="P"),c("GO_ID", "gene")]
go_obo<- read.csv("../GO_annotations/ontology_obo_out.csv")
go_obo<- plyr::rename(go_obo, replace=c("id"="GO_ID"))
go2name_bp<- go_obo[which(go_obo$namespace=="biological_process"),c("GO_ID", "name")]


## annotating the LOC no
genes_key<- read.csv("../GO_annotations/Maggies_annotations_modifiedR.csv")
genes_key<- plyr::rename(genes_key, replace=c("GeneID"="gene"))
genes_key$gene[genes_key$gene=="LOC113993669"] <- "CYP19A1"
genes_key$gene[genes_key$gene=="LOC113983511"] <- "OXT"
genes_key$gene[genes_key$gene=="LOC113983498"] <- "AVP"
genes_key$gene[genes_key$gene=="LOC113982601"] <- "AVPR2"
data_genes<- data.frame(gene=rownames(data))
genes_key<- merge(data_genes, genes_key, by="gene", all.x=TRUE)
#genes_key<- genes_key[which(grepl("LOC[0-9]+",genes_key$gene)),]

genes_key$display_gene_ID<- ifelse(grepl("LOC[0-9]+", genes_key$gene) & !grepl("LOC[0-9]+",genes_key$best_anno) & !is.na(genes_key$best_anno), paste0(genes_key$gene," (",genes_key$best_anno,")"), as.character(genes_key$gene))
genes_key<- genes_key[,c("gene","best_anno","display_gene_ID")]
  
candidates2<- c("AR", "SRD5A2", "CYP19A1", "ESR1", "ESR2", "PGR","GNRH1", "PRL","PRLR","VIP", "VIPR1", "VIPR2","OXT", "OXTR", "AVP", "AVPR1A", "AVPR1B", "AVPR2")

I have to run each tissue separately in part because of linear combinations of batch variables and tissues (some tissues were sequenced only on a single flow cell) that means DESeq2 throws a “model not full rank” error. But also we are interested in the within-tissue gene expression.



1 Gonads (GON)

I will use the first two tissues to demonstrate the processes applied across all tissues, but code echoing will be suppressed in future tissues so as to save space.

tissue="GON"
gon_key<- subset(key_behav, Tissue==tissue)
gon_key<- droplevels(gon_key)
gon_data<- data[,colnames(data) %in% rownames(gon_key)]


start<- nrow(gon_data)
#remove genes with less than 5 reads
gon_data$avg_count<- apply(gon_data, 1, mean)
gon_data<- gon_data[gon_data$avg_count>5,]
gon_data$avg_count<-NULL


#remove genes where >50% of samples have 0 gene expression
gon_data$percent_0<- apply(gon_data, 1, function(x)length(x[x==0]))
thresh<- ncol(gon_data)/2
gon_data<- gon_data[gon_data$percent_0<=thresh,]
gon_data$percent_0<-NULL

Before filtering we had 16854 genes, after filtering for mean read count and excluding genes where >50% of samples had a count of 0 we are left with 16169

1.1 Checking the sampling

Before I go into the analyses, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have gonad samples for 16, with respect to T and the tracking data we have 12 samples.

Here, we will run all the data to check for DEGs against status, then subset the data to only include those individuals with T and behavioural data.

fsb<- fisher.test(table(gon_key$Status, gon_key$Batch))
knitr::kable(table(gon_key$Status, gon_key$Batch), caption=paste0("Status across sequencing runs, fisher test p=", fsb$p.value)) %>% kable_styling()
Status across sequencing runs, fisher test p=1
pilot run1
floater 2 5
territorial 2 7
fsy<- fisher.test(table(gon_key$Status, gon_key$Year))
knitr::kable(table(gon_key$Status, gon_key$Year), caption=paste0("Status across sampling years, fisher test p=", fsy$p.value)) %>% kable_styling()
Status across sampling years, fisher test p=1
2015 2017 2018
floater 2 3 2
territorial 2 4 3

1.2 Status GON

First, I will run through a data checking process. I will check for outliers, and then find out how our potential nuisance variables effect our gene expression data. If they have effects they will be incorporated into the final DESeq2 models.


1.2.1 Expression data exploration

dd<- DESeqDataSetFromMatrix(countData=gon_data, colData=gon_key, design= ~ Status)
dd<- DESeq(dd)
dd<- dd[which(mcols(dd)$betaConv),] #remove any genes that didn't converge. 


plotDispEsts(dd)

#plot a PCA of our samples to look for suspicious samples and any obvious patterning with respect to our interest variables
vsd_data<- getVarianceStabilizedData(dd)

p <- pca(vsd_data, metadata = gon_key)
a<- biplot(p, lab=gon_key$Harvest_ID, colby="Status", shape="Batch", legendPosition="right", title="Gonads all samples PCA") + peri_figure
a

## check to see how our interest and nuisance variables are correlated to the PC axes. This will help decide on model form for DESeq.
b<- eigencorplot(p, metavars = c("time_kill","Batch","Year", "Status"))
FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"
b

#use WGCNA connectivity measure to identify outliers.

datExpr0<- as.data.frame(t(vsd_data))
#gsg<- goodSamplesGenes(datExpr0, verbose=3)
#gsg$allOK

A=adjacency(t(datExpr0),type="signed")
#-----Calculate whole network connectivity
k=as.numeric(apply(A,2,sum))-1
#-----Standardized connectivity
Z.k=scale(k)
thresholdZ.k=-2.5 
outlierColor=ifelse(Z.k<thresholdZ.k,"red","black")
sampleTree = flashClust(as.dist(1-A), method = "average")
#-----Convert traits to colors

datColors=data.frame(outlier=outlierColor)
#-----Plot the sample dendrogram
plotDendroAndColors(sampleTree,groupLabels=names(datColors),
                    colors=datColors,main="Sample dendrogram and trait heatmap")

Based on the eigencor plot, I will include Batch as a variable in my analyses, as it is associated with PC4.


1.2.1.1 Corrections applied

Also let’s see how the data look when we account for Batch in our model by using limma::RemoveBatchEffect()

#remove batch effect
mat<- limma::removeBatchEffect(vsd_data, gon_key$Batch)
p <- pca(mat, metadata = gon_key)
biplot(p, lab=gon_key$Harvest_ID, colby="Status", shape="Batch", legendPosition="right", title="PCA Gonads removeBatchEffect(Batch)")

#use this if Batch and Year are both pinged in the previous section. 
eigencorplot(p, metavars = c("Batch","Year", "Status"))
FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

wgcnadata<- as.data.frame(t(mat))


traits<- gon_key[, c("Batch",  "Year")]
traits$pilot<- as.numeric(ifelse(traits$Batch=="pilot",1,0))
traits$Run1<- as.numeric(ifelse(traits$Batch=="run1",1,0))
traits$Run2<- as.numeric(ifelse(traits$Batch=="run2",1,0))
traits$yr2015<- as.numeric(ifelse(traits$Year=="2015",1,0))
traits$yr2017<- as.numeric(ifelse(traits$Year=="2017",1,0))
traits$yr2018<- as.numeric(ifelse(traits$Year=="2018",1,0))

traits<- traits[,-(1:2)] 

A=adjacency(t(wgcnadata),type="signed")
#-----Calculate whole network connectivity
k=as.numeric(apply(A,2,sum))-1
#-----Standardized connectivity
Z.k=scale(k)
thresholdZ.k=-2.5 
outlierColor=ifelse(Z.k<thresholdZ.k,"red","black")
sampleTree = flashClust(as.dist(1-A), method = "average")
#-----Convert traits to colors
traitColors=data.frame(numbers2colors(traits,signed=FALSE))
dimnames(traitColors)[[2]]=paste(names(traits))
datColors=data.frame(outlier=outlierColor,traitColors)

#-----Plot the sample dendrogram
plotDendroAndColors(sampleTree,groupLabels=names(datColors),
                    colors=datColors,main="Sample dendrogram and trait heatmap")

1.2.2 Analysis

Now run DEseq with the updated model form and extract DEGs.

Formula ~ Batch + Status.

varname="Status"
design(dd)<- formula(~ Batch + Status)



dd<- DESeq(dd)

res<- results(dd, alpha=0.1)
res<- res[order(res$padj),]
summary(res)
FALSE 
FALSE out of 16169 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 2, 0.012%
FALSE LFC < 0 (down)     : 0, 0%
FALSE outliers [1]       : 18, 0.11%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 2)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results
par(mfrow=c(1,2))
DESeq2::plotMA(res, ylim = c(-2, 2), main=paste("MA-plot",tissue,"exp",design(dd)))
hist(res$pvalue, breaks=100, col="grey", main="", xlab="uncorrected P-value")

dev.off()
FALSE null device 
FALSE           1
des<- as.character(design(dd))
des<- paste0(des, collapse=" ")
out_res<- data.frame(res)
out_res$gene<- row.names(out_res)
out_res<- merge(out_res, genes_key,by="gene")
out_res<- out_res[order(out_res$pvalue),]
out_res$Tissue<- tissue
out_res$design<- des
out_res$fdrtool<- "no"

write.csv(out_res, file="../DE_results/results_GON_Status.csv", row.names=TRUE)

gene_plots(out_res,c("Status","Class"),6, status_cols)
g


1.2.3 Gene Ontology

Conduct Gene Set enrichment analysis (GSEA) when there are few significant differentially expressed genes. This will leverage the whole list. Then, I will also conduct foreground background testing to get an explicit view of either the top genes after FDR correction (q<0.1), or the top genes without FDR correction (p<0.05)

res_gsea_bp(out_res)

gseaplot

write.csv(gsearesult, file=paste0("../DE_results/GSEA_BP_",tissue,"_",varname,".csv"), row.names=FALSE)
res_go_bp(out_res)

go_plot_result

knitr::kable(enrich_results[1:10,-1], caption=paste0("Top 10 enriched GO BP in ",tissue, " according to ",varname)) %>% kable_styling()
Top 10 enriched GO BP in GON according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0010951 negative regulation of endopeptidase activity 21/817 79/12553 0.0000000 0.0000333 0.0000326 CPAMD8/LOC113992168/KNG1/SERPINB1/SERPINA10/SPP2/ITIH4/FETUB/LOC113991485/ITIH2/VTN/LOC113989405/C3/SERPINC1/SERPINF2/AGT/BIRC6/HRG/C5/USP14/SERPINF1 21 817
GO:0030449 regulation of complement activation 13/817 37/12553 0.0000003 0.0002476 0.0002422 C6/LOC113984930/LOC113992168/F2/C8B/VTN/C3/CFI/LOC113986226/C9/C5/C8A/CFH 13 817
GO:0042730 fibrinolysis 8/817 14/12553 0.0000007 0.0003848 0.0003765 FGB/F2/FGG/PLG/LOC113992335/SERPINF2/FGA/HRG 8 817
GO:0051918 negative regulation of fibrinolysis 7/817 11/12553 0.0000013 0.0004807 0.0004702 F2/PLG/LOC113992335/SERPINF2/APOH/THBD/HRG 7 817
GO:0002576 platelet degranulation 22/817 108/12553 0.0000014 0.0004807 0.0004702 CLEC3B/SPARC/FGB/FGG/KNG1/SPP2/ITIH4/ENDOD1/PLG/F5/ALB/LOC113992335/SERPINF2/FGA/APOH/LOC114001803/LOC113992533/HRG/TMX3/LOC113989827/TMSB4X/APOOL 22 817
GO:0007596 blood coagulation 24/817 128/12553 0.0000022 0.0006423 0.0006284 GATA5/ANO6/FGB/F2/FGG/SERPINA10/PROC/LOC113995989/F9/F7/PLG/F5/SERPINC1/LOC113992335/CAPZB/FGA/LOC113995424/THBD/IRF2/COL1A2/RCOR1/PRSS56/COL1A1/GATA2 24 817
GO:0006957 complement activation, alternative pathway 8/817 18/12553 0.0000076 0.0017094 0.0016724 LOC113984930/C8B/C3/LOC113986226/C9/C5/C8A/CFH 8 817
GO:0044267 cellular protein metabolic process 27/817 165/12553 0.0000078 0.0017094 0.0016724 STC2/LAMB2/CP/LOC113992168/F2/FGG/KNG1/SERPINA10/SPP2/PROC/GSN/LOC114003546/CCN1/PLG/ITIH2/F5/C3/SERPINC1/ALB/APOA5/LOC113992335/LOC113991601/FGA/TTR/MMP2/P4HB/SNCA 27 817
GO:0006958 complement activation, classical pathway 9/817 27/12553 0.0000323 0.0063096 0.0061727 C6/LOC113992168/C8B/C3/CFI/LOC113986226/C9/C5/C8A 9 817
GO:0019835 cytolysis 5/817 10/12553 0.0002207 0.0352323 0.0344678 C6/C8B/C9/C5/C8A 5 817
go_table<- enrich_results[1:10,c("ID","Description","GeneRatio","pvalue","p.adjust")]
go_table<- go_table[!is.na(go_table$ID),]
go_table$pvalue<- signif(go_table$pvalue,2)
go_table$p.adjust<- signif(go_table$p.adjust,2)

go_table<- ggtexttable(as.data.frame(go_table),theme = ttheme(base_size = 6,padding=unit(c(4,10),"pt")),rows=NULL) %>% tab_add_title(text = main.title, size=6, padding = unit(0.1, "line"))%>% tab_add_title(text = main.title, size=6, padding = unit(0.1, "line"))



write.csv(enrich_results, file=paste0("../DE_results/GO_BP_",tissue,"_",varname,".csv"), row.names=FALSE)

Below is the code for making the figures in the supplement.

g2<- ggarrange(ggarrange(vp,g,widths=c(0.7,1), labels=c("A","B"),ncol=2,nrow=1),ggarrange(gseaplot,go_table, ncol=2,nrow=1, labels=c("C","D")), nrow=2)


#g2
ggsave(filename="../DE_results/supp_figure_GON_status.png", plot=g2, device="png" ,height=180, width=200, units="mm", bg="white")


1.3 Testosterone GON

Given that the behaviors are best correlated with mean T, then this is our variable of interest


1.3.1 Expression data exploration

But first, we need to look at these new data in the gonads with the new subset with respect to our interest variables in the PCA.

rm(list= ls()[!(ls() %in% c(keep,"gon_data","gon_behav"))])
tissue="GON"
gon_behav<- subset(key_behav, Tissue==tissue)
gon_behav<- subset(gon_behav, Batch!="pilot")
gon_behav<- droplevels(gon_behav)
gon_data<- data[,colnames(data) %in% rownames(gon_behav)]

start<- nrow(gon_data)
#remove genes with less than 5 reads
gon_data$avg_count<- apply(gon_data, 1, mean)
gon_data<- gon_data[gon_data$avg_count>5,]
gon_data$avg_count<-NULL


#remove genes where >50% of samples have 0 gene expression
gon_data$percent_0<- apply(gon_data, 1, function(x)length(x[x==0]))
thresh<- ncol(gon_data)/2
gon_data<- gon_data[gon_data$percent_0<=thresh,]
gon_data$percent_0<-NULL

Before filtering we had 16854 genes, after filtering for mean read count and excluding genes where >50% of samples had a count of 0 we are left with 16137…

dd<- DESeqDataSetFromMatrix(countData=gon_data, colData=gon_behav, design= ~ mean_T)
dd<- DESeq(dd)
dd<- dd[which(mcols(dd)$betaConv),] #remove any genes that didn't converge. 
#plotDispEsts(dd)

vsd_data<- getVarianceStabilizedData(dd)
#vsd_data<- assay(vsd)

write.csv(counts(dd, normalized=TRUE), "../DE_results/data_GON_norm.csv", row.names=TRUE, quote=FALSE)

p <- pca(vsd_data, metadata = gon_behav)
biplot(p, lab=gon_behav$Harvest_ID, colby="mean_T", shape="Year", legendPosition="right", title="mean T in GON")

eigencorplot(p, metavars = c("time_kill","Year","Status", "mean_T","strength.all_study"))
FALSE [1] "Status"
FALSE [1] "Year"

datExpr0<- as.data.frame(t(vsd_data))
#gsg<- goodSamplesGenes(datExpr0, verbose=3)
#gsg$allOK

A=adjacency(t(datExpr0),type="signed")
#-----Calculate whole network connectivity
k=as.numeric(apply(A,2,sum))-1
#-----Standardized connectivity
Z.k=scale(k)
thresholdZ.k=-2.5 
outlierColor=ifelse(Z.k<thresholdZ.k,"red","black")
sampleTree = flashClust(as.dist(1-A), method = "average")
#-----Convert traits to colors

datColors=data.frame(outlier=outlierColor)
#-----Plot the sample dendrogram
plotDendroAndColors(sampleTree,groupLabels=names(datColors),
                    colors=datColors,main="Sample dendrogram and trait heatmap")


1.3.2 Corrections applied

FALSE [1] "Status"
FALSE [1] "Year"


1.3.3 Analysis

DESeq model specification ~ Year + mean_T

This chunk also includes a function I wrote to remove highly influential observations (as measured by Cook’s Distance) from the continuous data. I know from previous iterations of this analysis that there are observations that are highly influential and also DESeq2 does not exclude these from the results in continuous data.

varname="mean_T"
dd<- DESeqDataSetFromMatrix(countData=gon_data, colData=gon_behav, design= ~ Year + mean_T)
dd<- DESeq(dd)
dd<- dd[which(mcols(dd)$betaConv),]
res<- results(dd, alpha=0.1)
summary(res)
FALSE 
FALSE out of 16137 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 33, 0.2%
FALSE LFC < 0 (down)     : 111, 0.69%
FALSE outliers [1]       : 0, 0%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results
#DEseq2 doesn't do outlier removal for continuous variables. I'm doing that here with a bespoke function
res<- rm_continuous_outliers(dd,res)
res<- res[order(res$padj),]
summary(res)
FALSE 
FALSE out of 16137 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 33, 0.2%
FALSE LFC < 0 (down)     : 111, 0.69%
FALSE outliers [1]       : 37, 0.23%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results
des<- as.character(design(dd))
des<- paste0(des, collapse=" ")
out_res<- data.frame(res)
out_res$gene<- row.names(out_res)
out_res<- merge(out_res, genes_key,by="gene")
out_res<- out_res[order(out_res$pvalue),]
out_res$Tissue<- tissue
out_res$design<- des
out_res$fdrtool<- "no"

write.csv(out_res, file="../DE_results/results_GON_mean_T.csv", row.names=TRUE)



par(mfrow=c(1,2))
DESeq2::plotMA(res, ylim = c(-2, 2), main=paste("MA-plot",tissue,"exp",des))
hist(res$pvalue, breaks=100, col="grey", main="", xlab="uncorrected P-value")

dev.off()
FALSE null device 
FALSE           1


1.3.4 Gene Ontology

Top 10 enriched GO BP in GON according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0050729 positive regulation of inflammatory response 16/1105 57/12541 0.0000213 0.0305125 0.0298555 PDE2A/AGTR1/LOC114003559/IFI35/SNCA/LPL/LOC114000059/TNIP1/TLR2/TGM2/IL1RL1/CEBPB/PLA2G7/CLOCK/ETS1/APP 16 1105
GO:0051412 response to corticosterone 7/1105 13/12541 0.0000432 0.0305125 0.0298555 STAR/TH/AVPR1A/LOC113992757/STARD3/LOC113998857/NEFL 7 1105
GO:0050679 positive regulation of epithelial cell proliferation 15/1105 55/12541 0.0000555 0.0305125 0.0298555 NR4A3/LOC113991446/HTRA1/OSR2/LOC113991466/MYC/LOC113986457/BMP4/BMP5/FGFR2/IGF1/TWIST1/HRAS/FGF1/ATP7A 15 1105
GO:0006614 SRP-dependent cotranslational protein targeting to membrane 19/1105 81/12541 0.0000593 0.0305125 0.0298555 LOC113990217/RPL31/RPL37/RPS20/RPS25/RPL39/RPS12/RPL23/RPL38/RPL22/RPL35/RPL29/RPL32/RPL23A/RPS6/RPL9/RPL5/RPL27/RPL10A 19 1105
GO:0019083 viral transcription 20/1105 100/12541 0.0003898 0.1578854 0.1544856 LOC113990217/RPL31/RPL37/RPS20/RPS25/RPL39/RPS12/RPL23/RPL38/RPL22/RPL35/RPL29/RPL32/RPL23A/RPS6/RPL9/RPL5/TPR/RPL27/RPL10A 20 1105
GO:0006825 copper ion transport 6/1105 13/12541 0.0004603 0.1578854 0.1544856 ATOX1/HEPHL1/SLC31A1/FKBP4/ATP7A/CUTC 6 1105
GO:0055010 ventricular cardiac muscle tissue morphogenesis 7/1105 18/12541 0.0005396 0.1586287 0.1552129 ISL1/FOXC1/FGFR2/LOC113998261/FKBP1A/PTCD2/TPM1 7 1105
GO:0046928 regulation of neurotransmitter secretion 6/1105 14/12541 0.0007457 0.1918374 0.1877065 SNCG/SNCAIP/CPLX2/CPLX1/RHOT1/MCTP1 6 1105
GO:0021510 spinal cord development 8/1105 25/12541 0.0009806 0.1955375 0.1913269 BAG3/PTN/PLXDC1/SUZ12/PHGDH/SRD5A1/NEFL/PTPRS 8 1105
GO:0050772 positive regulation of axonogenesis 9/1105 31/12541 0.0010495 0.1955375 0.1913269 SHOX2/METRN/STK25/CHODL/NGF/PLXND1/MAP6/NEFL/SLITRK1 9 1105


1.4 Status x mean T GON

For pulling out candidate genes with interactions

rm(list= ls()[!(ls() %in% c(keep,"gon_data","gon_behav","tissue"))])
varname="Status x mean_T"
dd<- DESeqDataSetFromMatrix(countData=gon_data, colData=gon_behav, design= ~ Year + Status + mean_T + Status:mean_T)
dd<- DESeq(dd,test="LRT",reduced=~Year + Status + mean_T)
#dd<- DESeq(dd)
#dd<- dd[which(mcols(dd)$betaConv),]


#summary(res)
res<- results(dd, alpha=0.1)
res<- rm_continuous_outliers(dd,res)
res<- res[order(res$padj),]
summary(res)
FALSE 
FALSE out of 16137 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 5, 0.031%
FALSE LFC < 0 (down)     : 7, 0.043%
FALSE outliers [1]       : 29, 0.18%
FALSE low counts [2]     : 933, 5.8%
FALSE (mean count < 14)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results
des<- as.character(design(dd))
des<- paste0(des, collapse=" ")
out_res<- data.frame(res)
out_res$gene<- row.names(out_res)
out_res<- merge(out_res, genes_key,by="gene")
out_res<- out_res[order(out_res$padj),]
out_res$Tissue<- tissue
out_res$design<- des
out_res$n_p005<- nrow(out_res[which(out_res$pvalue<0.05),])
out_res$n_q01<- nrow(out_res[which(out_res$padj<0.1),])
out_res$Test<- "LRT_full_interaction"
out_res$fdrtool<- "no"

write.csv(out_res, file="../DE_results/interactions_GON_mean_T_Status.csv", row.names=FALSE)



par(mfrow=c(1,2))
DESeq2::plotMA(res, ylim = c(-2, 2), main=paste("MA-plot",tissue,"exp",des))
hist(res$pvalue, breaks=100, col="grey", main="", xlab="uncorrected P-value")

dev.off()
FALSE null device 
FALSE           1

How sensitive is the candidate gene results to model form?


1.5 Strength GON

1.5.1 Analysis

formula ~ Year + scale(Strength)

Without scaling strength, DESeq2 would throw a warning, recommending the variable be scaled.

rm(list= ls()[!(ls() %in% c(keep,"gon_data","gon_behav","tissue","dd"))])
varname="Strength"
dd$strength.all_study<- scale(dd$strength.all_study)
design(dd)<- formula(~Year + strength.all_study)
dd<- DESeq(dd)

res<- results(dd, alpha=0.1)

res<- rm_continuous_outliers(dd,res)
summary(res)
FALSE 
FALSE out of 16137 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 10, 0.062%
FALSE LFC < 0 (down)     : 11, 0.068%
FALSE outliers [1]       : 45, 0.28%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results
res<- res[order(res$padj),]

des<- as.character(design(dd))
des<- paste0(des, collapse=" ")
out_res<- data.frame(res)
out_res$gene<- row.names(out_res)
out_res<- merge(out_res, genes_key,by="gene")
out_res<- out_res[order(out_res$padj),]
out_res$Tissue<- tissue
out_res$design<- des
out_res$fdrtool<- "no"


write.csv(out_res, file="../DE_results/results_GON_Strength.csv", row.names=FALSE)


par(mfrow=c(1,2))
DESeq2::plotMA(res, ylim = c(-2, 2), main=paste("MA-plot",tissue,"exp",des))
hist(res$pvalue, breaks=100, col="grey", main="", xlab="uncorrected P-value")

dev.off()
FALSE null device 
FALSE           1


1.5.2 Gene Ontology

Top 10 enriched GO BP in GON according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0045540 regulation of cholesterol biosynthetic process 13/919 35/12541 0.0000005 0.0009732 0.0009732 SC5D/MVK/FDFT1/FASN/SEC14L2/ELOVL6/LOC113987727/HMGCS1/LOC114002832/MVD/ACACA/SREBF1/LSS 13 919
GO:0006695 cholesterol biosynthetic process 12/919 35/12541 0.0000038 0.0035789 0.0035789 MVK/FDFT1/MSMO1/DHCR24/LOC113987727/HMGCS1/LOC114002832/MVD/ACLY/APOA5/LSS/APOA1 12 919
GO:0044267 cellular protein metabolic process 28/919 164/12541 0.0000211 0.0112266 0.0112266 UBA52/SERPINA10/ADORA2A/LOC113992168/PLG/ITIH2/SERPINC1/F2/TTR/FN1/ALB/TGOLN2/SPP2/LOC113992658/FGG/APOA5/F5/FGA/IGFBP4/CALB1/SERPIND1/SHISA5/LOC113996623/GGA1/CCN1/FSTL3/APOA1/LOC113990797 28 919
GO:0042730 fibrinolysis 7/919 14/12541 0.0000241 0.0112266 0.0112266 PLG/HRG/F2/FGB/FGG/SERPINF2/FGA 7 919
GO:0010951 negative regulation of endopeptidase activity 16/919 79/12541 0.0001629 0.0534749 0.0534749 SERPINA10/VTN/LOC113992168/LOC113989405/ITIH2/HRG/SERPINC1/PTTG1/FETUB/SPOCK3/SPP2/SERPINF2/SERPIND1/GAPDH/ITIH4/AGT 16 919
GO:1904158 axonemal central apparatus assembly 7/919 18/12541 0.0001721 0.0534749 0.0534749 LOC113997903/LOC113997531/LOC113995484/LOC114001265/LOC113997354/LOC113983738/DNAJB13 7 919
GO:0070328 triglyceride homeostasis 7/919 19/12541 0.0002554 0.0680072 0.0680072 ABCG5/ANGPTL3/APOA5/ANGPTL4/GIP/LIPC/APOA1 7 919
GO:0006694 steroid biosynthetic process 7/919 20/12541 0.0003682 0.0857953 0.0857953 FDFT1/LOC114002832/LOC113992757/STARD3/STAR/LOC113983393/LSS 7 919
GO:0008652 cellular amino acid biosynthetic process 7/919 21/12541 0.0005177 0.0957789 0.0957789 OAT/PAH/ASPG/GLUL/GLS/GLUD1/ALDH18A1 7 919
GO:0060438 trachea development 6/919 16/12541 0.0006437 0.0957789 0.0957789 LOC113997903/LOC113997531/LOC113995484/LOC114001265/LOC113997354/LOC113983738 6 919



2 Pituitary (PIT)

2.1 Checking the sampling

Before filtering we had 16854 genes, after filtering for mean read count and excluding genes where >50% of samples had a count of 0 we are left with 14812

Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have pit samples for 12, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=1
run1 run2
floater 3 2
territorial 5 2
Sampling years distributed across batches, fisher test p=0.22
2017 2018
run1 6 2
run2 1 3


2.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"


2.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

Looks controlling for Batch also removes the association with Year. Let’s just include Batch.


2.2 Status PIT

2.2.1 Analysis

Formula = ~ Batch + Status

In this section, I find results with different p-value distributions. I was following THIS tutorial by Bernd Klaus and it recommended using fdrtool to correct these p-values.

varname="Status"
design(dd)<- formula(~ Batch + Status)
dd<- DESeq(dd)
res<- results(dd, alpha=0.1)


res1<- res[!is.na(res$padj),]
res1<- res1[!is.na(res1$pvalue),]
res1$padj<- NULL
res1.fdr<- fdrtool(res1$stat, statistic="normal", plot=FALSE, verbose=FALSE)
res1$padj<- p.adjust(res1.fdr$pval, method="BH")

outs<- res[is.na(res$padj),] # add those outliers back in... 
res1<- rbind(res1, outs)
res1<- res1[order(res1$padj),]
summary(res1)
FALSE 
FALSE out of 14812 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 8, 0.054%
FALSE LFC < 0 (down)     : 6, 0.041%
FALSE outliers [1]       : 91, 0.61%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 4)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results
des<- as.character(design(dd))
des<- paste0(des, collapse=" ")
out_res<- data.frame(res1)
out_res$gene<- row.names(out_res)
out_res<- merge(out_res, genes_key,by="gene")
out_res<- out_res[order(out_res$padj),]
out_res$Tissue<- tissue
out_res$design<- des
out_res$fdrtool<- "yes"

write.csv(out_res, file="../DE_results/results_PIT_status.csv", row.names=FALSE)

par(mfrow=c(2,2))
DESeq2::plotMA(res, ylim = c(-2, 2), main=paste("MA-plot",tissue,"exp",des))
hist(res$pvalue, breaks=20, col="grey", main="", xlab="uncorrected P-value")
DESeq2::plotMA(res1, ylim = c(-2, 2), main="FDR tool corrected PIT Status")
hist(res1.fdr$pval, breaks=20, col="grey", main="P-value correction 'fdrtool'", xlab="'Corrected' P-value")

dev.off()
FALSE null device 
FALSE           1

2.2.2 Gene Ontology


Top 10 enriched GO BP in PIT according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0007422 peripheral nervous system development 7/426 23/12076 0.0000097 0.0125810 0.0123228 LOC113999850/SOX10/SERPINI1/LOC113986199/PMP22/NTF3/GFRA3 7 426
GO:0051384 response to glucocorticoid 8/426 41/12076 0.0000770 0.0498862 0.0488623 PAM/ALPL/SDC1/LOC113993087/ADIPOQ/CASP7/TYMS/SPARC 8 426
GO:0032496 response to lipopolysaccharide 12/426 100/12076 0.0002011 0.0752137 0.0736699 GCH1/ALPL/FMO1/CASP7/PCK1/CNR1/DCN/VCAM1/OPRM1/SPARC/CNP/LIAS 12 426
GO:0045776 negative regulation of blood pressure 5/426 17/12076 0.0002323 0.0752137 0.0736699 GCH1/ADIPOQ/RNLS/CNR1/GJA5 5 426
GO:0010951 negative regulation of endopeptidase activity 9/426 73/12076 0.0010203 0.2106709 0.2063470 SERPIND1/ITIH2/LOC113991485/SERPINF1/COL6A3/SERPINF2/SERPINI1/ITIH4/LOC113992706 9 426
GO:0006957 complement activation, alternative pathway 4/426 14/12076 0.0011533 0.2106709 0.2063470 CFH/C8B/C7/LOC113986241 4 426
GO:0042552 myelination 7/426 47/12076 0.0012002 0.2106709 0.2063470 ADGRG6/MPZ/LOC114000755/PLLP/DHH/MAL2/BCAS1 7 426
GO:0006836 neurotransmitter transport 5/426 24/12076 0.0013014 0.2106709 0.2063470 SLC22A3/LOC113990781/SLC6A2/SLC29A1/SLC6A4 5 426
GO:0071300 cellular response to retinoic acid 7/426 51/12076 0.0019578 0.2445362 0.2395173 SERPINF1/BRINP2/EPHA3/OSR1/PCK1/ADNP2/SLC6A4 7 426
GO:0007200 phospholipase C-activating G protein-coupled receptor signaling pathway 6/426 38/12076 0.0019694 0.2445362 0.2395173 PTGER3/ESR1/P2RY2/CMKLR1/HTR2B/OPRM1 6 426

2.3 Testosterone PIT

2.3.1 Analysis

FALSE 
FALSE out of 14812 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 14, 0.095%
FALSE LFC < 0 (down)     : 5, 0.034%
FALSE outliers [1]       : 62, 0.42%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 4)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


2.3.2 Gene Ontology

Top 10 enriched GO BP in PIT according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0010759 positive regulation of macrophage chemotaxis 4/470 13/12076 0.0012238 0.5854897 0.5782776 LOC113985892/IL34/CSF1/LOC113983017 4 470
GO:0045087 innate immune response 20/470 245/12076 0.0014440 0.5854897 0.5782776 LOC114003556/ALPK1/IL34/PGLYRP2/SUSD4/INAVA/LOC113996654/CSF1/LOC113992821/SARM1/LOC113992189/LOC113983017/HEXIM1/SRC/NOD1/SLC15A4/S100B/LOC113986435/STYK1/SYK 20 470
GO:0002009 morphogenesis of an epithelium 4/470 14/12076 0.0016610 0.5854897 0.5782776 LOC113990794/SOX10/DAG1/TIMELESS 4 470
GO:0046849 bone remodeling 4/470 14/12076 0.0016610 0.5854897 0.5782776 P3H4/LTBP3/WNT16/TPH1 4 470
GO:0006813 potassium ion transport 7/470 48/12076 0.0023882 0.6235461 0.6158653 KCNG1/KCNA4/KCNN1/LOC113993556/KCNJ15/KCNAB1/KCNS2 7 470
GO:0007015 actin filament organization 11/470 109/12076 0.0034092 0.6235461 0.6158653 SHROOM4/LOC114003556/SORBS3/CORO6/ERMN/MARCKSL1/LOC113992312/ELMO1/LOC113988218/SAMD14/TMSB15B 11 470
GO:2000134 negative regulation of G1/S transition of mitotic cell cycle 5/470 27/12076 0.0034699 0.6235461 0.6158653 GFI1B/DACT1/CDKN1A/CDKN2C/EZH2 5 470
GO:0032755 positive regulation of interleukin-6 production 7/470 52/12076 0.0037979 0.6235461 0.6158653 LOC114003556/LOC114000755/INAVA/IL6R/NOD1/LOC113986435/SYK 7 470
GO:0030335 positive regulation of cell migration 16/470 197/12076 0.0043787 0.6235461 0.6158653 LOC113991284/MYLK/DAPK3/HBEGF/LOC113996864/PLP1/CSF1/CXCL12/CTSH/LOC113983017/BMP2/TMSB15B/PAK1/SMO/NTRK3/FLT1 16 470
GO:0006337 nucleosome disassembly 4/470 18/12076 0.0044873 0.6235461 0.6158653 LOC113985909/LOC113985906/SMARCC2/LOC113997381 4 470


2.4 mean T x Status

FALSE 
FALSE out of 14812 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 1205, 8.1%
FALSE LFC < 0 (down)     : 1401, 9.5%
FALSE outliers [1]       : 37, 0.25%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 4)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


2.5 Strength PIT

2.5.1 Analysis

Formula: ~ Batch + scale(Strength)

FALSE 
FALSE out of 14812 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 387, 2.6%
FALSE LFC < 0 (down)     : 113, 0.76%
FALSE outliers [1]       : 207, 1.4%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 4)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1

As you can see above, PFT3 has consistently high leverage on the social network analysis. Even if I play around with the outlier theshold this I am always getting non-linear plots like above. So, let’s see what happens when I remove this sample.


2.5.2 Re-analysis - outlier removed

FALSE 
FALSE out of 14812 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 109, 0.74%
FALSE LFC < 0 (down)     : 13, 0.088%
FALSE outliers [1]       : 58, 0.39%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 2)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


2.5.3 Gene Ontology

Top 10 enriched GO BP in PIT according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0019934 cGMP-mediated signaling 7/1174 15/12076 0.0002572 0.3249333 0.3247639 PDE2A/NPR1/NPR2/EDNRB/LOC113993563/HTR2C/AQP1 7 1174
GO:0043116 negative regulation of vascular permeability 7/1174 16/12076 0.0004190 0.3249333 0.3247639 PDE2A/CLDN5/ANGPT4/DDAH1/ADORA2A/LOC113996612/ARHGAP35 7 1174
GO:0009566 fertilization 7/1174 17/12076 0.0006528 0.3249333 0.3247639 TDRD9/BAX/AAAS/LOC113990833/TDRKH/SYCP2/YBX3 7 1174
GO:0034097 response to cytokine 11/1174 38/12076 0.0007119 0.3249333 0.3247639 LOC114001824/MAPKAPK2/MCL1/TIMP2/LOC113992821/BCL2L1/IL6R/LOC113984425/ALDH1A2/TNFRSF11A/TYMS 11 1174
GO:0046777 protein autophosphorylation 28/1174 155/12076 0.0009200 0.3249333 0.3247639 STK11/MAP3K11/CAD/MAPKAPK2/DAPK3/CHEK2/LOC114000515/PIM1/MARK2/MKNK2/CAMK2B/VRK2/FGFR1/ACVR1B/LOC113983017/STK39/CDK5/MAP3K12/LOC113994408/DYRK1A/CSNK1G2/LOC113999057/ERBB2/DDR2/FER/EIF2AK2/NTRK1/ATP13A2 28 1174
GO:0016525 negative regulation of angiogenesis 17/1174 77/12076 0.0009651 0.3249333 0.3247639 PRL/NPR1/PTPRM/AGBL5/CLDN5/ADGRB2/HSPG2/ISM1/LOC113993563/EPHA2/SEMA3E/SPRY2/KLF4/GADD45A/THBS4/FASLG/HGS 17 1174
GO:1902430 negative regulation of amyloid-beta formation 6/1174 14/12076 0.0012596 0.3562365 0.3560508 LOC113988015/LOC113987965/SORL1/CHRNA7/PIN1/PRNP 6 1174
GO:0030259 lipid glycosylation 5/1174 10/12076 0.0014289 0.3562365 0.3560508 ST3GAL2/ST3GAL4/B3GALT1/LOC114003933/GBGT1 5 1174
GO:0007015 actin filament organization 21/1174 109/12076 0.0016920 0.3562365 0.3560508 SAMD14/LOC113988218/LOC113992312/CORO1B/SORBS3/INPPL1/KPTN/LOC114003556/TPM4/TMOD1/CORO6/MYO1D/SHROOM4/FAT1/RHOH/HIP1/SHROOM2/LMOD1/HIP1R/MYO5C/RHOB 21 1174
GO:0070527 platelet aggregation 11/1174 42/12076 0.0017635 0.3562365 0.3560508 HSPB1/ITGA2B/MYL9/STXBP1/COMP/FIBP/PEAR1/LOC114002136/RAP2B/ITGB3/METAP1 11 1174



3 Ventromedial Hypothalamus (VMH)

3.1 Checking the sampling

Before filtering we had 16854 genes, after filtering for mean read count and excluding genes where >50% of samples had a count of 0 we are left with 15220

Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have vmh samples for 16, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=0.76
pilot run1 run2
floater 2 4 1
territorial 2 7 0
Sampling years distributed across batches, fisher test p=0.00078
2015 2017 2018
pilot 4 0 0
run1 0 6 5
run2 0 1 0

The majority of our data in VMH is in run 1.

3.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

3.1.2 Corrections applied

FALSE [1] "Status"
FALSE [1] "Year"

Looks like controlling for Batch is sufficient here, as Batch is highly correlated with Year.

3.2 Status VMH

3.2.1 Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 15220 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 0, 0%
FALSE LFC < 0 (down)     : 3, 0.02%
FALSE outliers [1]       : 4, 0.026%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


3.2.2 Gene Ontology

Top 10 enriched GO BP in VMH according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0032981 mitochondrial respiratory chain complex I assembly 10/615 49/12283 0.0001285 0.1376210 0.1376210 ACAD9/NDUFA5/NDUFS7/TIMM21/NDUFA10/NDUFS3/NDUFV2/NDUFB9/NDUFA8/NDUFAF3 10 615
GO:0006614 SRP-dependent cotranslational protein targeting to membrane 13/615 81/12283 0.0001773 0.1376210 0.1376210 RPS15A/RPS23/RPS29/RPL27A/RPL35A/RPL19/RPS21/RPS24/UBA52/LOC113989206/RPS6/RPS8/RPL6 13 615
GO:0000184 nuclear-transcribed mRNA catabolic process, nonsense-mediated decay 14/615 112/12283 0.0013789 0.4774939 0.4774939 RPS15A/RPS23/RPS29/RPL27A/RPL35A/RPL19/RPS21/RPS24/UBA52/PNLDC1/LOC113989206/RPS6/RPS8/RPL6 14 615
GO:0019083 viral transcription 13/615 100/12283 0.0014081 0.4774939 0.4774939 RPS15A/RPS23/RPS29/RPL27A/RPL35A/RPL19/RPS21/RPS24/UBA52/LOC113989206/RPS6/RPS8/RPL6 13 615
GO:0051604 protein maturation 5/615 18/12283 0.0015383 0.4774939 0.4774939 TSPAN5/LOC114002160/STUB1/TSPAN15/TSPAN33 5 615
GO:0006413 translational initiation 14/615 119/12283 0.0024636 0.5339842 0.5339842 RPS15A/RPS23/RPS29/RPL27A/RPL35A/RPL19/RPS21/RPS24/UBA52/EIF3J/LOC113989206/RPS6/RPS8/RPL6 14 615
GO:0006120 mitochondrial electron transport, NADH to ubiquinone 7/615 38/12283 0.0024770 0.5339842 0.5339842 NDUFA5/NDUFS7/NDUFA10/NDUFS3/NDUFV2/NDUFB9/NDUFA8 7 615
GO:0006412 translation 16/615 148/12283 0.0030068 0.5339842 0.5339842 IGF2BP3/RPS15A/RPS23/RPS29/RPL27A/RPL35A/RPL19/RPS21/MRPL13/RPS24/GGT7/LOC113989206/RMND1/RPS6/RPS8/RPL6 16 615
GO:0070207 protein homotrimerization 4/615 13/12283 0.0030966 0.5339842 0.5339842 CLYBL/LOC113994845/LOC113996480/LOC113989423 4 615
GO:0031145 anaphase-promoting complex-dependent catabolic process 9/615 65/12283 0.0048481 0.7524203 0.7524203 UBB/ANAPC7/PSMF1/ANAPC2/PSMB7/PSMB4/UBA52/LOC113998129/PSMC5 9 615


3.3 Testosterone VMH


3.3.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"


3.3.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

Batch does not account for year here… so have to include both…


3.3.3 Analysis

Formula: ~ Batch + Year + mean_T

FALSE 
FALSE out of 15162 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 3, 0.02%
FALSE LFC < 0 (down)     : 8, 0.053%
FALSE outliers [1]       : 52, 0.34%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


3.3.4 Gene Ontology

Top 10 enriched GO BP in VMH according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0030198 extracellular matrix organization 22/514 209/12262 0.0000660 0.0918544 0.0918544 VIT/COL9A1/COL1A1/COL7A1/PDGFRA/COL5A2/ADAMTS14/CCDC80/ITGB3/FBLN5/COLQ/NR2E1/LOC113997171/COL4A2/ADAMTS7/FN1/TGFBI/ADAMTSL5/LOC113998180/LAMB2/LOC113982147/LOC113993281 22 514
GO:0060285 cilium-dependent cell motility 5/514 13/12262 0.0001235 0.0918544 0.0918544 DNAH1/DNAH12/RFX3/CCDC65/LOC113984044 5 514
GO:0007611 learning or memory 8/514 50/12262 0.0010246 0.4615527 0.4615527 GRIN2A/FZD9/S100B/LHX8/MME/CASP3/PSEN1/PTPRZ1 8 514
GO:0090023 positive regulation of neutrophil chemotaxis 5/514 21/12262 0.0014776 0.4615527 0.4615527 LOC113988506/LOC113985892/DNM1L/LOC113988345/RIPOR2 5 514
GO:0007601 visual perception 15/514 153/12262 0.0019106 0.4615527 0.4615527 MFRP/COL1A1/LOC113999869/LOC113993535/NDP/PDE6G/LOC114004117/LOC113987867/NR2E1/ZIC2/SLC24A1/TGFBI/GLRB/LAMB2/LOC113982931 15 514
GO:0042542 response to hydrogen peroxide 7/514 44/12262 0.0021723 0.4615527 0.4615527 COL1A1/DUSP1/STARD3/CASP3/FKBP1B/STK26/PRDX3 7 514
GO:0002244 hematopoietic progenitor cell differentiation 7/514 45/12262 0.0024815 0.4615527 0.4615527 LOC114000366/PDGFRA/PTPRC/TENT2/LOC113997171/PSEN1/SIN3A 7 514
GO:0060041 retina development in camera-type eye 7/514 45/12262 0.0024815 0.4615527 0.4615527 MFRP/TENT2/LOC114004117/LOC113987867/NR2E1/LAMB2/LOC113982931 7 514
GO:0036159 inner dynein arm assembly 4/514 15/12262 0.0028794 0.4760682 0.4760682 LOC113981955/DNAH1/DNAH12/TEKT2 4 514
GO:0001947 heart looping 7/514 48/12262 0.0036142 0.5377997 0.5377997 NDRG4/ZIC3/WNT5A/VANGL2/TMED2/PSEN1/SHH 7 514


3.4 Status x meanT

FALSE 
FALSE out of 15162 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 10, 0.066%
FALSE LFC < 0 (down)     : 16, 0.11%
FALSE outliers [1]       : 141, 0.93%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


3.5 Strength VMH

3.5.1 Analysis

Formula ~ Batch + Year + scale(Strength)

FALSE 
FALSE out of 15162 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 28, 0.18%
FALSE LFC < 0 (down)     : 19, 0.13%
FALSE outliers [1]       : 82, 0.54%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


3.5.2 Gene Ontology

Top 10 enriched GO BP in VMH according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0042472 inner ear morphogenesis 5/174 40/12262 0.0002391 0.1730775 0.1643205 FGFR2/COL2A1/USH1C/FOXG1/KCNQ4 5 174
GO:0006488 dolichol-linked oligosaccharide biosynthetic process 3/174 15/12262 0.0011271 0.2719956 0.2582337 ALG1/SRD5A3/ALG13 3 174
GO:0036159 inner dynein arm assembly 3/174 15/12262 0.0011271 0.2719956 0.2582337 DNAH12/LOC113981955/CCDC39 3 174
GO:0007605 sensory perception of sound 7/174 126/12262 0.0021050 0.3234432 0.3070782 LOC113990629/COL1A1/OTOG/COL2A1/USH1C/KCNQ4/ATP6V0A4 7 174
GO:0003341 cilium movement 4/174 40/12262 0.0024034 0.3234432 0.3070782 CFAP61/DNAH12/LOC113981955/CCDC39 4 174
GO:0010765 positive regulation of sodium ion transport 3/174 20/12262 0.0026805 0.3234432 0.3070782 LOC113986459/LOC113988585/LOC113999252 3 174
GO:0021537 telencephalon development 3/174 23/12262 0.0040363 0.3331375 0.3162820 LOC113986459/LOC113988585/RFX4 3 174
GO:0050679 positive regulation of epithelial cell proliferation 4/174 52/12262 0.0062423 0.3331375 0.3162820 LOC113986459/LOC113988585/FGFR2/WDR48 4 174
GO:0000077 DNA damage checkpoint 3/174 29/12262 0.0078260 0.3331375 0.3162820 DOT1L/CLSPN/ERCC6 3 174
GO:0000076 DNA replication checkpoint 2/174 10/12262 0.0083599 0.3331375 0.3162820 CLSPN/TIPIN 2 174



4 Anterior Hypothalamus (AH)


4.1 Checking the sampling

For the AH, we started with 16854 genes but after filering we have 14926 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have ah samples for 11, with respect to T and the tracking data.

Status across sequencing runs
run2
floater 4
territorial 7
Sampling years distributed across batches
2017 2018
run2 6 5


4.1.1 Expression data exploration

FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE


PFT1 and PFT11 cluster with the BSTm as previously identified by the brain tissue PCA plot and the all tissue WGCNA dendrogram. We know that there were issues with the sectioning plane in individual PFT1 that could affect this tissue, but not PFT11.

The effects of removing these samples will be explored for each analysis vs removing a randomly selected pair of samples. These results (not shown) showed similar effects, so these samples were not excluded.


4.2 Status AH

4.2.1 Analysis

Formula = ~ Status

FALSE null device 
FALSE           1


4.2.2 Gene Ontology

Top 10 enriched GO BP in AH according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0071850 mitotic cell cycle arrest 7/1048 14/12134 0.0000701 0.1385595 0.1385595 DUSP1/CDC14A/RGCC/CDKN1A/DDIAS/CDC14B/FAP 7 1048
GO:0051017 actin filament bundle assembly 9/1048 29/12134 0.0005283 0.5222015 0.5222015 DPYSL3/USH1C/ADD2/AIF1L/CALD1/ESPN/EZR/FAM107A/SHROOM2 9 1048
GO:0000226 microtubule cytoskeleton organization 20/1048 109/12134 0.0009542 0.6288193 0.6288193 LOC113996267/LOC113991233/TUBB3/WEE1/MAPT/TACC1/CDK5/CDK5RAP2/SON/APC2/CDC14A/LOC113995052/TBCD/TUBA8/TUBG1/GAS2/SIK3/TBCEL/HAUS6/CDC14B 20 1048
GO:0010389 regulation of G2/M transition of mitotic cell cycle 15/1048 75/12134 0.0016430 0.8120696 0.8120696 LOC113996267/LOC113984333/CCP110/ALMS1/LOC113999915/DCTN2/CDK5RAP2/PRKAR2B/LOC113995052/NEDD1/PPP2R1A/TUBG1/LOC113986004/TPX2/HAUS6 15 1048
GO:0051604 protein maturation 6/1048 18/12134 0.0030639 0.9303018 0.9303018 TSPAN33/LMF2/TSPAN14/GALNT2/TSPAN32/LMF1 6 1048
GO:0000278 mitotic cell cycle 19/1048 113/12134 0.0035755 0.9303018 0.9303018 LOC113996267/RRM1/CLTC/LOC113991233/TUBB3/LOC114001298/CDC14A/LOC113995052/NUDT15/TBCD/TUBA8/TADA3/TUBG1/TPX2/ABRAXAS2/LOC113997232/YEATS4/CDC14B/WAPL 19 1048
GO:0070509 calcium ion import 7/1048 25/12134 0.0042182 0.9303018 0.9303018 LOC113989062/MCUR1/LOC113994017/CDK5/LOC114000517/LOC113993166/SLC30A1 7 1048
GO:0000086 G2/M transition of mitotic cell cycle 20/1048 127/12134 0.0060890 0.9303018 0.9303018 LOC113996267/LOC113984333/CCP110/USH1C/WEE1/ALMS1/FBXL17/LOC113999915/DCTN2/CDK7/CDK5RAP2/PRKAR2B/LOC113995052/NEDD1/PPP2R1A/TUBG1/CDKN1A/LOC113986004/TPD52L1/HAUS6 20 1048
GO:0051895 negative regulation of focal adhesion assembly 6/1048 21/12134 0.0071579 0.9303018 0.9303018 LOC113988419/LOC113991284/SRC/BCAS3/FAM107A/LRP1 6 1048
GO:0016485 protein processing 11/1048 56/12134 0.0075717 0.9303018 0.9303018 NCSTN/LOC113991284/ECEL1/PCSK7/AEBP1/MMEL1/ECE2/FURIN/ADAM10/PCSK6/PSEN1 11 1048


4.3 Testosterone AH

4.3.1 Analysis

Formula ~ mean_T

FALSE 
FALSE out of 14926 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 1, 0.0067%
FALSE LFC < 0 (down)     : 2, 0.013%
FALSE outliers [1]       : 33, 0.22%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


4.3.2 Gene Ontology

Top 10 enriched GO BP in AH according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0042572 retinol metabolic process 8/632 32/12134 0.0001789 0.2810090 0.2810090 ALDH1A3/LOC113995092/LOC113996821/LOC113999436/LOC114001898/LOC114001962/PNPLA2/RETSAT 8 632
GO:0072205 metanephric collecting duct development 4/632 10/12134 0.0011897 0.5788714 0.5788714 CALB1/DLG5/LOC114001962/SHH 4 632
GO:0006414 translational elongation 5/632 17/12134 0.0013827 0.5788714 0.5788714 ABTB1/EEF2/GTPBP1/LOC113991380/TSFM 5 632
GO:0042552 myelination 8/632 45/12134 0.0020169 0.5788714 0.5788714 ACSBG1/ATRN/CMTM8/EGR2/LOC114001565/LOC114003174/NFASC/SERINC5 8 632
GO:0009617 response to bacterium 8/632 46/12134 0.0023324 0.5788714 0.5788714 C3/GPM6A/LOC113995953/LPL/LSM5/NAALADL2/RAB29/UPK1B 8 632
GO:0061640 cytoskeleton-dependent cytokinesis 5/632 19/12134 0.0023826 0.5788714 0.5788714 NOX5/LOC113992196/SEPT10/SEPT2/SEPT4 5 632
GO:0030301 cholesterol transport 4/632 12/12134 0.0025793 0.5788714 0.5788714 APOA1/CD36/LCAT/OSBPL5 4 632
GO:0043691 reverse cholesterol transport 4/632 13/12134 0.0035736 0.7017611 0.7017611 APOA1/APOC3/LCAT/LOC113997963 4 632
GO:0072089 stem cell proliferation 4/632 14/12134 0.0047992 0.7821617 0.7821617 ABCB5/CD34/FGF2/RNF43 4 632
GO:0042110 T cell activation 5/632 23/12134 0.0058025 0.7821617 0.7821617 CD44/KIF13B/PPP3CA/PPP3CB/RAB29 5 632


4.4 Status x mean T

FALSE 
FALSE out of 14926 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 1220, 8.2%
FALSE LFC < 0 (down)     : 1340, 9%
FALSE outliers [1]       : 32, 0.21%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


4.5 Strength AH

4.5.1 Analysis

Formula ~ scale(Strength)

FALSE 
FALSE out of 14900 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 3, 0.02%
FALSE LFC < 0 (down)     : 1, 0.0067%
FALSE outliers [1]       : 0, 0%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


4.5.2 Gene Ontology

Top 10 enriched GO BP in AH according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0008217 regulation of blood pressure 9/478 58/12115 0.0004087 0.5554226 0.5554226 LOC113990682/NPR2/ATP1A1/GCGR/GCH1/MC3R/ACTA2/LOC113993563/EXTL1 9 478
GO:0090305 nucleic acid phosphodiester bond hydrolysis 8/478 52/12115 0.0009033 0.6137750 0.6137750 RAD9B/PLSCR1/GEN1/RBBP8/ASTE1/ENPP1/ENPP2/LOC114001323 8 478
GO:0042493 response to drug 19/478 226/12115 0.0015653 0.7091026 0.7091026 DAB1/SFRP2/LOC114000060/LOC114000287/MCM7/ATP1A1/WFDC1/ABCB5/GNAO1/GGH/APOD/LOC113989520/TP73/SORD/ADIPOR2/ATR/AGPAT2/PAM/SREBF1 19 478
GO:0006911 phagocytosis, engulfment 4/478 17/12115 0.0037808 0.7170805 0.7170805 ARHGAP12/DOCK1/MFGE8/GSN 4 478
GO:0051000 positive regulation of nitric-oxide synthase activity 4/478 17/12115 0.0037808 0.7170805 0.7170805 NOS1AP/PIK3CB/GCH1/S100A1 4 478
GO:0006814 sodium ion transport 7/478 52/12115 0.0040939 0.7170805 0.7170805 SLC5A3/SLC17A8/SLC12A3/SLC5A11/SLC13A4/SLC38A3/NDUFA9 7 478
GO:0006027 glycosaminoglycan catabolic process 5/478 28/12115 0.0043368 0.7170805 0.7170805 IDS/GLB1L2/SDC1/SGSH/FUCA1 5 478
GO:0007009 plasma membrane organization 5/478 28/12115 0.0043368 0.7170805 0.7170805 LOC113987015/LOC113993146/TLCD2/LOC113991140/LOC113993758 5 478
GO:0061025 membrane fusion 6/478 42/12115 0.0057369 0.7170805 0.7170805 LOC113987015/LOC113993146/LOC113991140/LOC113993758/VAMP3/NSFL1C 6 478
GO:0034341 response to interferon-gamma 3/478 10/12115 0.0059520 0.7170805 0.7170805 SLC11A1/CALCOCO2/GCH1 3 478



5 Paraventricular Nucleus (PVN)


5.1 Checking the sampling

For the PVN, we started with 16854 genes but after filering we have 15151 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have pvn samples for 12, with respect to T and the tracking data.

Status across sequencing runs
run2
floater 5
territorial 7

5.1.1 Expression data exploration

FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

Looks like we don’t have any outliers, nor do we need to include Year (or Batch) as a covariate.


5.2 Status PVN

5.2.1 Analysis

Formula = ~ Status

FALSE 
FALSE out of 15151 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 27, 0.18%
FALSE LFC < 0 (down)     : 15, 0.099%
FALSE outliers [1]       : 19, 0.13%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 2)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


5.2.2 Gene Ontology

Top 10 enriched GO BP in PVN according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0003341 cilium movement 11/447 38/12245 0.0000001 0.0000898 0.0000853 DNAH1/TTC25/LOC113981955/CFAP73/LOC113986306/DNAAF1/DNAI1/LOC114000425/LOC114000433/TTC29/ARMC4 11 447
GO:0010951 negative regulation of endopeptidase activity 12/447 70/12245 0.0000075 0.0050385 0.0047860 ITIH5/PROS1/A2ML1/ANXA2/LOC113992168/CRIM1/CD109/LOC113991485/LOC113989407/LOC113992817/SERPINB5/WFDC1 12 447
GO:0006695 cholesterol biosynthetic process 8/447 34/12245 0.0000233 0.0104184 0.0098962 FDFT1/DHCR7/DHCR24/HMGCS1/LOC113987727/APOA1/INSIG1/ACLY 8 447
GO:0045540 regulation of cholesterol biosynthetic process 7/447 34/12245 0.0001883 0.0631879 0.0600208 FDFT1/DHCR7/HMGCS1/SREBF2/LOC113987727/SC5D/SREBF1 7 447
GO:1904659 glucose transmembrane transport 5/447 17/12245 0.0002725 0.0731458 0.0694797 SLC5A3/MFSD4A/SLC2A12/LOC113984297/SLC5A1 5 447
GO:0007288 sperm axoneme assembly 5/447 20/12245 0.0006236 0.1394892 0.1324978 DNAH1/CFAP65/CFAP44/NEURL1B/MEIG1 5 447
GO:0032570 response to progesterone 6/447 31/12245 0.0007745 0.1484808 0.1410387 LOC113997556/TGFB2/CAV1/LOC113992301/SREBF1/GABRB1 6 447
GO:0071711 basement membrane organization 4/447 13/12245 0.0009634 0.1616051 0.1535052 CAV1/EXTL1/NID1/RIC8A 4 447
GO:0032094 response to food 4/447 14/12245 0.0013101 0.1758144 0.1670023 SREBF1/LOC113992180/OPRM1/CPS1 4 447
GO:0048566 embryonic digestive tract development 4/447 14/12245 0.0013101 0.1758144 0.1670023 LOC113997556/TGFB2/RARRES2/ALDH1A2 4 447


5.3 Testosterone PVN

5.3.1 Analysis

Formula = ~ mean_T

FALSE 
FALSE out of 15151 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 7, 0.046%
FALSE LFC < 0 (down)     : 6, 0.04%
FALSE outliers [1]       : 0, 0%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


5.3.2 Gene Ontology

Top 10 enriched GO BP in PVN according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0034587 piRNA metabolic process 4/399 11/12245 0.0003055 0.3676509 0.3553683 MAEL/PIWIL1/FKBP6/HENMT1 4 399
GO:0007186 G protein-coupled receptor signaling pathway 30/399 484/12245 0.0005584 0.3676509 0.3553683 POMC/PRLH/C3/MLNR/LOC113987091/PDGFRL/CYSLTR1/HTR2A/TACR3/ADGRV1/LOC113996140/LOC113997278/HRH3/FZD8/LOC114000970/GPR171/C3AR1/ADGRF4/PTHLH/HEBP1/FGD2/RASD1/LOC113993299/GRP/MCF2L/GPR6/LOC113999236/LOC114003556/ANXA1/LPAR1 30 399
GO:0021542 dentate gyrus development 4/399 14/12245 0.0008573 0.3676509 0.3553683 BTG2/EMX2/PROX1/ATAT1 4 399
GO:0050900 leukocyte migration 10/399 97/12245 0.0012129 0.3676509 0.3553683 LOC113984271/CD34/LOC113984305/GPC1/COL1A1/GRB2/ITGB3/SHC1/CD44/LOC113989681 10 399
GO:0048705 skeletal system morphogenesis 5/399 27/12245 0.0015989 0.3676509 0.3553683 SPEF2/LOC113985828/HHIP/FMN1/LOC113996506 5 399
GO:0006911 phagocytosis, engulfment 4/399 17/12245 0.0018862 0.3676509 0.3553683 LOC113984271/LOC113984305/LOC113992914/GULP1 4 399
GO:0051591 response to cAMP 5/399 28/12245 0.0018951 0.3676509 0.3553683 JUN/DUSP1/FOS/COL1A1/LOC114003020 5 399
GO:0035924 cellular response to vascular endothelial growth factor stimulus 5/399 29/12245 0.0022296 0.3784718 0.3658277 NR4A1/MAP2K3/FLT4/SPRY2/ANXA1 5 399
GO:0042127 regulation of cell population proliferation 10/399 107/12245 0.0025434 0.3819735 0.3692124 JUN/PLA2G4A/NFKBIA/LOC113983017/LAMA5/JUND/SIX3/TGFB2/EPS15L1/LOC114003556 10 399
GO:0035865 cellular response to potassium ion 3/399 10/12245 0.0034734 0.3819735 0.3692124 CRHBP/PPARGC1A/LOC113983406 3 399


5.4 Status x mean T

FALSE 
FALSE out of 15151 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 1, 0.0066%
FALSE LFC < 0 (down)     : 2, 0.013%
FALSE outliers [1]       : 13, 0.086%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


5.5 Strength PVN

5.5.1 Analysis

Formula ~ scale(Strength)

FALSE 
FALSE out of 15151 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 2, 0.013%
FALSE LFC < 0 (down)     : 3, 0.02%
FALSE outliers [1]       : 29, 0.19%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


5.5.2 Gene Ontology

Top 10 enriched GO BP in PVN according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0016082 synaptic vesicle priming 8/1044 20/12245 0.0001349 0.2591926 0.2591926 LOC113993146/SNAP25/LOC113987015/LOC113991140/LOC113993758/LOC114001826/LOC113986517/STXBP1 8 1044
GO:0016070 RNA metabolic process 11/1044 47/12245 0.0016303 0.8810927 0.8810927 HNRNPA0/DDX54/DUSP11/HNRNPA2B1/HNRNPA3/HNRNPU/PCBP2/POLR2C/PTBP1/NCBP1/HNRNPK 11 1044
GO:0120009 intermembrane lipid transfer 9/1044 36/12245 0.0026239 0.8810927 0.8810927 ABCA1/ABCA3/PITPNM2/PITPNC1/PLEKHA3/TTPA/OSBP/PITPNM1/PRELID3A 9 1044
GO:0006342 chromatin silencing 6/1044 18/12245 0.0028693 0.8810927 0.8810927 MECP2/EED/HMGB1/PCGF2/MORF4L1/LOC113997379 6 1044
GO:0006959 humoral immune response 7/1044 24/12245 0.0030460 0.8810927 0.8810927 RBPJ/LOC113992699/LOC114001683/NOTCH1/NOTCH2/TNFRSF21/LOC114001688 7 1044
GO:0030316 osteoclast differentiation 7/1044 24/12245 0.0030460 0.8810927 0.8810927 TCIRG1/CSF1R/LOC113996964/SBNO2/LOC114001688/GAB2/OSTM1 7 1044
GO:0090026 positive regulation of monocyte chemotaxis 5/1044 13/12245 0.0032106 0.8810927 0.8810927 CREB3L1/CREB3L2/MOSPD2/HMGB1/LOC113991763 5 1044
GO:0051402 neuron apoptotic process 9/1044 38/12245 0.0038941 0.9026353 0.9026353 XIAP/ATN1/BNIP3/BOK/NCSTN/SCN1A/TNFRSF21/LOC113994464/BID 9 1044
GO:0010812 negative regulation of cell-substrate adhesion 5/1044 14/12245 0.0046473 0.9026353 0.9026353 COL1A1/ANGPT2/CORO2B/NOTCH1/DMTN 5 1044
GO:0007568 aging 20/1044 126/12245 0.0048296 0.9026353 0.9026353 FBXO4/CLDN1/ATG7/ENDOG/PAX2/EEF2/RPN2/CTNNA1/GCLM/GNAO1/HTR2A/LOC114000703/NUDT1/PENK/PITX3/PPARGC1A/VCAM1/WRN/ITGB2/NAPEPLD 20 1044



6 Medial Preoptic Area (POM)

6.1 Checking the sampling

For the POM, we started with 16854 genes but after filering we have 15318 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have pom samples for 12, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=1
pilot run1
floater 2 5
territorial 2 7
Sampling years distributed across batches, fisher test p=0.00055
2015 2017 2018
pilot 4 0 0
run1 0 7 5


6.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"


6.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

Don’t need to control for year because accounting for Batch seems to fix that.


6.1.3 Outlier removed

PFT2 was previously identified as a potential outlier, and it’s coming out quite separately here, even after all the corrections for batch effects. So, this one I am going to REMOVE.


##Status POM

###Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 15314 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 3, 0.02%
FALSE LFC < 0 (down)     : 2, 0.013%
FALSE outliers [1]       : 16, 0.1%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


6.1.4 Gene Ontology

Top 10 enriched GO BP in POM according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0042391 regulation of membrane potential 14/628 82/12335 0.0000602 0.0967256 0.0954846 GABRA4/LOC113991128/LOC113990721/GRIN1/CHRNB4/GABRA5/CHRNA7/KCNH7/KCNH8/LOC113984913/RIMS4/CHRNA5/KCNA1/KCNH6 14 628
GO:0071679 commissural neuron axon guidance 5/628 12/12335 0.0001977 0.1589313 0.1568921 PTCH1/NFIB/ROBO3/VEGFA/NRP1 5 628
GO:0006816 calcium ion transport 11/628 70/12335 0.0007607 0.4077390 0.4025075 ATP2B2/LOC113993165/GRIN3B/LOC113994017/CACNA1D/CHRNA7/LOC113982108/PSEN2/ITPR1/CACNA1E/CACNB3 11 628
GO:0071805 potassium ion transmembrane transport 13/628 99/12335 0.0014905 0.5329363 0.5260984 SLC9A3/CCDC51/KCNA4/KCND3/KCNH7/KCNH8/KCNK12/KCNK18/LOC113984913/KCNA1/KCNH6/KCNK15/SLC24A2 13 628
GO:0008589 regulation of smoothened signaling pathway 5/628 18/12335 0.0016571 0.5329363 0.5260984 PTCH1/INTU/IFT81/C2CD3/OTX2 5 628
GO:0050877 nervous system process 7/628 39/12335 0.0031741 0.6453285 0.6370485 GABRA4/LOC113991128/LOC113990721/CHRNB4/GABRA5/CHRNA7/CHRNA5 7 628
GO:0021756 striatum development 4/628 13/12335 0.0032905 0.6453285 0.6370485 CNTNAP2/DRD2/BBS4/SECISBP2 4 628
GO:0007268 chemical synaptic transmission 20/628 203/12335 0.0035735 0.6453285 0.6370485 GABRA4/SV2B/LOC113993165/LOC113991128/SLC18A2/HTR5A/SDCBP2/LOC113990721/LOC113994017/GRIN1/CHRNB4/GABRA5/CHRNA7/LOC113982386/SYN2/CHRM4/CHRNA5/KCNA1/CACNA1E/CACNB3 20 628
GO:0060384 innervation 4/628 14/12335 0.0044231 0.6453285 0.6370485 VCAM1/SULF2/GABRA5/SERPINE2 4 628
GO:0060749 mammary gland alveolus development 4/628 14/12335 0.0044231 0.6453285 0.6370485 AR/SOCS2/VEGFA/LOC113998275 4 628


6.2 Testosterone POM

6.2.1 Expression data exploration

FALSE [1] "Status"
FALSE [1] "Year"


6.2.2 Analysis

Formula = ~ mean_T

FALSE 
FALSE out of 15268 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 34, 0.22%
FALSE LFC < 0 (down)     : 11, 0.072%
FALSE outliers [1]       : 29, 0.19%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


6.2.3 Gene Ontology

Top 10 enriched GO BP in POM according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0070593 dendrite self-avoidance 6/493 20/12310 0.0000960 0.1208173 0.1201815 VSTM2L/MYPN/CNTN2/ROBO4/NEXN/ROBO3 6 493
GO:0010575 positive regulation of vascular endothelial growth factor production 6/493 22/12310 0.0001726 0.1208173 0.1201815 C3/FLT4/BRCA1/IL6ST/RORA/CCBE1 6 493
GO:0007186 G protein-coupled receptor signaling pathway 35/493 488/12310 0.0006065 0.2830267 0.2815371 PRLH/RGS8/MC5R/LHCGR/OPRL1/C3/GNRH1/GPR50/GPR171/RAC2/FGD3/MCF2L/RXFP3/GAL/SST/LOC113993299/CCL28/IAPP/RAPGEFL1/TSHR/LOC113989272/HRH3/SORT1/LOC114002880/GPR78/LOC113982386/SSTR1/P2RY6/HTR5A/LOC114004322/NTSR1/RIC8A/NPY/TIAM2/LOC113996140 35 493
GO:0007565 female pregnancy 8/493 52/12310 0.0009956 0.3484516 0.3466176 PAPPA/GNRH1/CRHBP/CORIN/LOC113989272/FOS/THBD/LOC113982386 8 493
GO:0046849 bone remodeling 4/493 14/12310 0.0018461 0.3793144 0.3773180 NOTCH2/TPH1/P3H4/RASSF2 4 493
GO:0007399 nervous system development 22/493 279/12310 0.0019248 0.3793144 0.3773180 PLPPR1/CPLX2/GDPD4/NOTCH2/NRN1L/IGF2BP3/ZC3H12A/GPSM1/HES4/CAMK1D/LOC113991778/GDPD5/TTLL7/MTR/VEGFA/SMARCD1/PCDH7/RAPGEFL1/FOS/BPNT1/PRPS1/NR2E1 22 493
GO:0007187 G protein-coupled receptor signaling pathway, coupled to cyclic nucleotide second messenger 7/493 46/12310 0.0021843 0.3793144 0.3773180 MC5R/LHCGR/TSHR/HRH3/SSTR1/HTR5A/NPY 7 493
GO:0006958 complement activation, classical pathway 5/493 24/12310 0.0022802 0.3793144 0.3773180 LOC113992168/C3/LOC113984305/LOC113984271/CFI 5 493
GO:0002064 epithelial cell development 4/493 15/12310 0.0024384 0.3793144 0.3773180 LOC113992388/LOC114000433/LOC113987257/TP63 4 493
GO:0033628 regulation of cell adhesion mediated by integrin 4/493 17/12310 0.0039897 0.5430211 0.5401631 TESC/PTK2/LOC114002160/LYN 4 493


6.3 mean T x Status

FALSE 
FALSE out of 15268 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 16, 0.1%
FALSE LFC < 0 (down)     : 8, 0.052%
FALSE outliers [1]       : 13, 0.085%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


6.4 Strength POM

6.4.1 Analysis

Formula ~ scale(Strength)

FALSE 
FALSE out of 15268 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 5, 0.033%
FALSE LFC < 0 (down)     : 10, 0.065%
FALSE outliers [1]       : 48, 0.31%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


6.4.2 Gene Ontology

Top 10 enriched GO BP in POM according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0043312 neutrophil degranulation 41/699 375/12310 0.0000408 0.0694043 0.0686333 HPSE/GNS/CD59/LOC114003217/TCIRG1/FUCA1/PSMD14/RAB18/TOLLIP/RAB7A/YPEL5/LOC113990017/SVIP/AP1M1/FGL2/GRN/FUCA2/SLC27A5/CRISPLD2/CTSB/SNAP25/APRT/COMMD9/LOC113992610/LTA4H/PRDX4/RAB24/HSP90AB1/PGAM1/PSMC3/ARHGAP45/RAP2C/LOC114002464/GUSB/PSMD12/PTPRC/PAFAH1B2/ACTR10/CREG1/DEGS1/LOC113996661 41 699
GO:0030574 collagen catabolic process 7/699 27/12310 0.0006035 0.3493887 0.3455077 MRC2/FURIN/ADAMTS14/MMP23B/CTSB/MMP15/LOC113997326 7 699
GO:0016082 synaptic vesicle priming 6/699 20/12310 0.0006399 0.3493887 0.3455077 LOC113987015/LOC113991140/LOC113993758/LOC113993146/LOC114002475/SNAP25 6 699
GO:0030010 establishment of cell polarity 8/699 37/12310 0.0009222 0.3493887 0.3455077 TCIRG1/SDCCAG8/RAB11FIP2/ARHGEF11/HSP90AB1/KIF26B/LOC113998953/CRK 8 699
GO:0016079 synaptic vesicle exocytosis 6/699 23/12310 0.0014401 0.3493887 0.3455077 LOC113987015/LOC113991140/LOC113993758/LOC113993146/SNAP25/CPLX1 6 699
GO:1900006 positive regulation of dendrite development 5/699 16/12310 0.0015037 0.3493887 0.3455077 OPA1/LRP8/LOC113989207/CRTC1/PTN 5 699
GO:2001224 positive regulation of neuron migration 5/699 16/12310 0.0015037 0.3493887 0.3455077 PLAA/LOC113989207/SEMA3A/LOC113989827/LOC114003556 5 699
GO:0019079 viral genome replication 4/699 10/12310 0.0016432 0.3493887 0.3455077 KPNA6/HTATSF1/CTBP2/LAMTOR5 4 699
GO:0033628 regulation of cell adhesion mediated by integrin 5/699 17/12310 0.0020317 0.3839828 0.3797175 LYN/LOC114002160/TESC/LOC113998953/CRK 5 699
GO:0051014 actin filament severing 4/699 12/12310 0.0035355 0.5622677 0.5560220 SCIN/VILL/CFL2/SVIL 4 699



7 Dorsomedial Intercollicular nucleus (ICo)


7.1 Checking the sampling

For the ICO, we started with 16854 genes but after filering we have 15069 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have ico samples for 11, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=0.41
pilot run1 run2
floater 1 1 3
territorial 2 5 2
Sampling years distributed across batches, fisher test p=0.015
2015 2017 2018
pilot 3 0 0
run1 0 4 2
run2 0 3 2

7.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

7.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE


7.2 Status ICO

7.2.1 Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 15069 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 2, 0.013%
FALSE LFC < 0 (down)     : 1, 0.0066%
FALSE outliers [1]       : 6, 0.04%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 4)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


7.2.2 Gene Ontology

Top 10 enriched GO BP in ICO according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0006614 SRP-dependent cotranslational protein targeting to membrane 15/532 81/12198 0.0000018 0.0025620 0.0025620 LOC114002081/RPS3/RPL22/RPL23A/RPL27/RPL37/RPL38/RPL4/RPL6/RPL7A/RPS14/RPS29/RPS4X/RPS6/UBA52 15 532
GO:0019083 viral transcription 16/532 100/12198 0.0000062 0.0043095 0.0043095 LOC114002081/RPS3/NUP160/RPL22/RPL23A/RPL27/RPL37/RPL38/RPL4/RPL6/RPL7A/RPS14/RPS29/RPS4X/RPS6/UBA52 16 532
GO:0000184 nuclear-transcribed mRNA catabolic process, nonsense-mediated decay 15/532 111/12198 0.0000915 0.0426901 0.0426901 LOC114002081/RPS3/RPL22/RPL23A/RPL27/RPL37/RPL38/RPL4/RPL6/RPL7A/RPS14/RPS29/RPS4X/RPS6/UBA52 15 532
GO:0006413 translational initiation 15/532 118/12198 0.0001840 0.0643631 0.0643631 LOC114002081/RPS3/RPL22/RPL23A/RPL27/RPL37/RPL38/RPL4/RPL6/RPL7A/RPS14/RPS29/RPS4X/RPS6/UBA52 15 532
GO:0045540 regulation of cholesterol biosynthetic process 7/532 34/12198 0.0005556 0.1554559 0.1554559 DHCR7/FDFT1/GGPS1/LPCAT3/LSS/SC5D/SREBF1 7 532
GO:0015949 nucleobase-containing small molecule interconversion 6/532 26/12198 0.0007299 0.1701948 0.1701948 AK2/AK9/CMPK1/GSR/LOC113996758/RRM1 6 532
GO:0006412 translation 15/532 148/12198 0.0020197 0.3846192 0.3846192 LOC114002081/RPS3/RPL22/RPL23A/RPL27/RPL37/RPL38/RPL4/RPL6/RPL7A/RPS14/RPS29/RPS4X/RPS6/TNIP1 15 532
GO:0006936 muscle contraction 10/532 79/12198 0.0021994 0.3846192 0.3846192 GAMT/TNNI1/CACNA1S/CLCN1/ITGA1/LOC114001935/MYL6/MYL9/TPM2/TTN 10 532
GO:0007076 mitotic chromosome condensation 4/532 15/12198 0.0033248 0.4651398 0.4651398 AKAP8L/LOC113992565/SMC4/TTN 4 532
GO:0070542 response to fatty acid 4/532 15/12198 0.0033248 0.4651398 0.4651398 CAT/FOXO1/FOXO3/LOC113996085 4 532


7.3 Testosterone ICO

7.3.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE


7.3.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

Still need to control for Year here…


7.3.3 Analysis

Formula = ~ Batch + Year + mean_T

FALSE 
FALSE out of 14968 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 121, 0.81%
FALSE LFC < 0 (down)     : 40, 0.27%
FALSE outliers [1]       : 16, 0.11%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


7.3.4 Gene Ontology

Top 10 enriched GO BP in ICO according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0003341 cilium movement 17/473 38/12155 0.00e+00 0.0000000 0.0000000 LOC113981955/DNAI1/DNAI2/DNAH9/LOC113998190/CFAP70/DNAH1/LOC114000425/ARMC4/LOC114000433/LRRC6/DNAH12/CFAP221/LOC114001265/CCDC40/TTC25/DNAH5 17 473
GO:0060285 cilium-dependent cell motility 8/473 14/12155 0.00e+00 0.0000079 0.0000077 LOC113984044/DNAH1/DNAH3/CFAP44/DNAH12/LOC113982904/EFHC1/EFHC2 8 473
GO:0030317 flagellated sperm motility 10/473 35/12155 6.00e-07 0.0002391 0.0002348 CFAP65/DNAI1/LOC113998190/DNAH1/LRRC6/ROPN1L/TCTE1/CCDC40/DNAH5/ENKUR 10 473
GO:0007368 determination of left/right symmetry 11/473 50/12155 2.60e-06 0.0007342 0.0007210 ZIC3/DNAI1/DNAI2/LOC113998190/LOC113984044/LOC113992699/FOXJ1/ARMC4/BBS7/LOC113982904/DNAH5 11 473
GO:0036158 outer dynein arm assembly 7/473 18/12155 2.80e-06 0.0007342 0.0007210 DNAI1/DNAI2/LOC113998190/ARMC4/LRRC6/TTC25/DNAH5 7 473
GO:0035082 axoneme assembly 7/473 21/12155 9.30e-06 0.0019436 0.0019086 CFAP74/JHY/LOC113992699/FOXJ1/CFAP46/CCDC40/RSPH1 7 473
GO:0007018 microtubule-based movement 12/473 68/12155 1.05e-05 0.0019436 0.0019086 DNAI1/DNAI2/DNAH9/LOC113998190/DNAH1/DNAH3/KIF24/LOC114000425/LOC114000985/DNAH12/DNAH10/DNAH5 12 473
GO:0007601 visual perception 18/473 148/12155 1.74e-05 0.0028142 0.0027635 GUCA1B/GUCA1C/NDP/ZIC2/COL1A1/LAMB2/LOC113984118/PDE6G/SOX14/USH2A/MYO3A/BBS7/MFRP/RORB/HSF4/NR2E1/GJC1/LOC113999235 18 473
GO:0018149 peptide cross-linking 5/473 12/12155 5.51e-05 0.0079406 0.0077974 F13A1/TGM1/LOC113988333/TGM2/LOC113992706 5 473
GO:0003351 epithelial cilium movement involved in extracellular fluid movement 5/473 13/12155 8.67e-05 0.0112426 0.0110399 LOC113998190/SPAG17/ROPN1L/CCDC40/DNAH5 5 473


7.4 mean T x Status

FALSE 
FALSE out of 14968 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 1194, 8%
FALSE LFC < 0 (down)     : 1086, 7.3%
FALSE outliers [1]       : 15, 0.1%
FALSE low counts [2]     : 1446, 9.7%
FALSE (mean count < 18)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


7.5 Strength ICO

7.5.1 Analysis

Formula ~ Batch + Year + scale(Strength)

FALSE 
FALSE out of 14968 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 39, 0.26%
FALSE LFC < 0 (down)     : 29, 0.19%
FALSE outliers [1]       : 11, 0.073%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


7.5.2 Gene Ontology

Top 10 enriched GO BP in ICO according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0007605 sensory perception of sound 15/306 127/12155 0.0000007 0.0007127 0.0006741 LOC113987015/LOC113993758/LOC113991140/BARHL1/ALDH7A1/COL4A3/GABRA5/ATP2B2/EPS8L2/USP53/LOC114000755/COL1A1/LRP2/LOC113993146/COL2A1 15 306
GO:0003341 cilium movement 7/306 38/12155 0.0000386 0.0201160 0.0190272 DNAI2/LOC114000425/DNAI1/LOC113981955/CFAP221/DNAH9/CCDC40 7 306
GO:0070268 cornification 6/306 28/12155 0.0000572 0.0201160 0.0190272 LOC114001273/DSG2/DSP/LOC114000789/LOC114002150/LOC113991144 6 306
GO:0007601 visual perception 13/306 148/12155 0.0000919 0.0234574 0.0221877 GUCA1B/GRM8/RPE65/GJC1/COL18A1/GLRA1/CNGB1/COL1A1/COL2A1/BBS4/LAMB2/LOC113999235/PAX6 13 306
GO:0016082 synaptic vesicle priming 5/306 20/12155 0.0001112 0.0234574 0.0221877 LOC113987015/LOC113993758/LOC113991140/LOC113993146/UNC13C 5 306
GO:0031424 keratinization 5/306 22/12155 0.0001812 0.0318573 0.0301329 LOC114001273/DSG2/DSP/LOC114000789/LOC114002150 5 306
GO:0030198 extracellular matrix organization 15/306 208/12155 0.0002497 0.0376358 0.0355986 COL4A4/COL4A3/LOC113993281/ADAMTS7/COL8A2/COL7A1/COL18A1/VCAN/COL1A1/COL8A1/COL2A1/COL4A6/COL6A2/LAMB2/LAMA3 15 306
GO:0043651 linoleic acid metabolic process 4/306 16/12155 0.0005642 0.0700335 0.0662427 LOC113991921/LOC114001775/LOC114002601/LOC114002603 4 306
GO:0035987 endodermal cell differentiation 5/306 28/12155 0.0005974 0.0700335 0.0662427 LOC113983959/COL7A1/COL8A1/COL4A6/LAMA3 5 306
GO:0007155 cell adhesion 19/306 331/12155 0.0007230 0.0762806 0.0721516 LOC113983959/FERMT2/SORBS3/CNTNAP2/COL4A3/CD9/CDHR4/LOC113988059/LOC113995955/LOC113993281/DSG2/PGM5/COL7A1/COL18A1/VCAN/COL8A1/COL6A2/STAB1/CX3CL1 19 306



8 Midbrain Central Grey (GCt)


8.1 Checking the sampling

For the GCT, we started with 16854 genes but after filering we have 15099 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have gct samples for 12, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=0.76
pilot run1 run2
floater 1 3 2
territorial 0 5 2
Sampling years distributed across batches, fisher test p=0.18
2015 2017 2018
pilot 1 0 0
run1 0 4 4
run2 0 3 1

given there is only 1 GCT sample from the pilot, I will just exclude the pilot data.


8.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"


8.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE


8.2 Status GCT

8.2.1 Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 15099 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 17, 0.11%
FALSE LFC < 0 (down)     : 12, 0.079%
FALSE outliers [1]       : 14, 0.093%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


8.2.2 Gene Ontology

Top 10 enriched GO BP in GCT according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0042738 exogenous drug catabolic process 4/294 10/12236 0.0000612 0.0428880 0.0407714 LOC113983773/NOS1/LOC114002604/LOC113988755 4 294
GO:0042572 retinol metabolic process 6/294 31/12236 0.0000811 0.0428880 0.0407714 LOC114001898/LOC113983773/LOC113988755/LOC113991249/BCO1/DGAT2 6 294
GO:0007160 cell-matrix adhesion 9/294 79/12236 0.0001113 0.0428880 0.0407714 NID1/COL3A1/FN1/ADAMTS12/ITGB5/NPNT/FGG/ITGA2/NID2 9 294
GO:0071356 cellular response to tumor necrosis factor 8/294 82/12236 0.0007711 0.2115720 0.2011302 CLDN1/LOC114000824/CRHBP/KLF2/ADAMTS12/NPNT/EDN1/LOC114003556 8 294
GO:0030522 intracellular receptor signaling pathway 5/294 32/12236 0.0009151 0.2115720 0.2011302 LOC113996216/LOC113995289/SREBF1/LOC113982931/NR2E1 5 294
GO:0030198 extracellular matrix organization 13/294 213/12236 0.0019262 0.3205493 0.3047290 NID1/VIT/COL3A1/LAMB2/LOC113999927/FN1/ADAMTS12/ITGB5/NPNT/FGG/ITGA2/NID2/NR2E1 13 294
GO:0006805 xenobiotic metabolic process 6/294 57/12236 0.0023716 0.3205493 0.3047290 LOC113983773/LOC114002604/LOC113996216/LOC113995289/LOC113988755/LOC113991029 6 294
GO:0006629 lipid metabolic process 8/294 98/12236 0.0024487 0.3205493 0.3047290 LCAT/SC5D/LOC113983773/LOC113998978/SREBF1/LOC113991249/CPNE7/DGAT2 8 294
GO:0018149 peptide cross-linking 3/294 12/12236 0.0025714 0.3205493 0.3047290 COL3A1/FN1/F13A1 3 294
GO:0043085 positive regulation of catalytic activity 4/294 25/12236 0.0027729 0.3205493 0.3047290 LOC114000824/FGFR4/SLC5A3/DCP1B 4 294


8.3 Testosterone GCT

8.3.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE null device 
FALSE           1


8.3.2 Analysis

Formula = ~ Batch + mean_T

FALSE 
FALSE out of 15105 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 4, 0.026%
FALSE LFC < 0 (down)     : 13, 0.086%
FALSE outliers [1]       : 33, 0.22%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


8.3.3 Gene Ontology

Top 10 enriched GO BP in GCT according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0007596 blood coagulation 14/380 115/12237 0.0000126 0.0170537 0.0160700 FGG/GATA2/F13A1/ENTPD1/LOC114003293/COL1A1/VWF/P2RX3/LOC113995989/SERPIND1/MICAL1/CDC42/ANO6/P2RX4 14 380
GO:0032703 negative regulation of interleukin-2 production 6/380 21/12237 0.0000315 0.0213293 0.0200990 LOC113986241/CD34/LOC114003293/LOC113993091/LOC113993083/PTPRC 6 380
GO:0006958 complement activation, classical pathway 6/380 23/12237 0.0000556 0.0250948 0.0236473 LOC113986241/C3/LOC113984271/LOC113984305/LOC113992168/C1R 6 380
GO:0045766 positive regulation of angiogenesis 12/380 105/12237 0.0000977 0.0330929 0.0311841 AQP1/APELA/GATA2/PRKD1/CD34/C3/ITGB2/EPHA1/WNT5A/VEGFC/NR2E1/ISL1 12 380
GO:2000178 negative regulation of neural precursor cell proliferation 4/380 10/12237 0.0001657 0.0448999 0.0423101 BTG2/GATA2/LIMS2/NR2E1 4 380
GO:0035735 intraciliary transport involved in cilium assembly 7/380 40/12237 0.0002026 0.0457568 0.0431175 HSPB11/DYNLL1/DYNC2LI1/KIF17/TTC26/IFT74/CLUAP1 7 380
GO:0006631 fatty acid metabolic process 8/380 55/12237 0.0002726 0.0527687 0.0497250 GPAM/HACL1/ACSBG2/C3/LOC113995989/TH/AASDH/LYPLA2 8 380
GO:0007411 axon guidance 17/380 212/12237 0.0003323 0.0562798 0.0530335 EMB/CXCR4/LOC113999057/PRKCQ/CSF1R/LOC114003293/EPHA1/ARX/MAPK7/LOC113988477/GBX2/WNT5A/SEMA3C/LYPLA2/BSG/NPTN/TNR 17 380
GO:0006957 complement activation, alternative pathway 4/380 14/12237 0.0007154 0.0866905 0.0816901 LOC113986241/C3/LOC113993091/LOC113993083 4 380
GO:0050776 regulation of immune response 6/380 36/12237 0.0007623 0.0866905 0.0816901 CD34/C3/ITGB2/COL1A1/CD99/COL3A1 6 380


8.4 mean T x Status

FALSE 
FALSE out of 15105 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 206, 1.4%
FALSE LFC < 0 (down)     : 189, 1.3%
FALSE outliers [1]       : 23, 0.15%
FALSE low counts [2]     : 583, 3.9%
FALSE (mean count < 9)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


8.5 Strength GCT

8.5.1 Analysis

Formula ~ Batch + scale(Strength)

FALSE 
FALSE out of 15105 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 34, 0.23%
FALSE LFC < 0 (down)     : 29, 0.19%
FALSE outliers [1]       : 41, 0.27%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


8.5.2 Gene Ontology

Top 10 enriched GO BP in GCT according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0006614 SRP-dependent cotranslational protein targeting to membrane 23/1035 81/12237 0.0000001 0.0002586 0.0002547 RPL39/RPL14/RPS24/RPL10A/RPSA/RPL34/RPS7/RPS8/RPS20/RPL31/RPL5/RPS3/LOC113987193/RPS25/RPL12/RPS15A/SSR3/RPL35A/RPS3A/RPL35/RPS5/RPS4X/RPL17 23 1035
GO:0006413 translational initiation 27/1035 119/12237 0.0000016 0.0015486 0.0015254 RPL39/RPL14/RPS24/RPL10A/EIF4A2/RPSA/RPL34/PABPC1/RPS7/RPS8/RPS20/RPL31/EIF4E3/RPL5/LOC113985040/RPS3/LOC113987193/RPS25/RPL12/RPS15A/RPL35A/RPS3A/RPL35/LOC114003727/RPS5/RPS4X/RPL17 27 1035
GO:0006412 translation 30/1035 148/12237 0.0000052 0.0033240 0.0032743 RPL39/IGF2BP3/RPL14/RPS24/RPL10A/RPSA/PSTK/RPL34/CPEB3/RPS7/RPS8/RPS20/MRPL18/RPL31/RSL24D1/RPL5/RPS3/LOC113987193/RPS25/RPL12/RPS15A/RPL35A/RPS3A/RPL35/LOC114003727/RPS5/MRPL55/RPS4X/GGT7/RPL17 30 1035
GO:0019083 viral transcription 23/1035 100/12237 0.0000074 0.0035429 0.0034898 RPL39/RPL14/RPS24/RPL10A/RPSA/RPL34/RPS7/RPS8/RPS20/RPL31/RPL5/RPS3/LOC113987193/RPS25/RPL12/RPS15A/RPL35A/RPS3A/RPL35/RAE1/RPS5/RPS4X/RPL17 23 1035
GO:0000184 nuclear-transcribed mRNA catabolic process, nonsense-mediated decay 24/1035 112/12237 0.0000170 0.0065104 0.0064129 RPL39/RPL14/RPS24/RPL10A/RPSA/RPL34/PABPC1/RPS7/RPS8/RPS20/RPL31/RPL5/RPS3/LOC113987193/RPS25/RPL12/RPS15A/RPL35A/RPS3A/RPL35/LOC114003727/RPS5/RPS4X/RPL17 24 1035
GO:0070050 neuron cellular homeostasis 7/1035 14/12237 0.0000612 0.0195424 0.0192497 TAOK1/NELL2/SLC6A2/MAP1A/CX3CL1/PRKAA1/LOC113995995 7 1035
GO:0030198 extracellular matrix organization 35/1035 213/12237 0.0001016 0.0277843 0.0273682 COL5A2/MMP2/ITGA2B/SH3PXD2A/LOC113982345/TNC/DAG1/LOC113991284/THBS1/HAPLN1/VTN/MMP15/ADAMTS3/ITGB2/LOC113998785/ADAMTS14/LUM/CCDC80/COL5A1/ACAN/SMOC2/TGFBI/MMP23B/COL6A3/PRDX4/ADAMTSL5/ELN/LAMA5/LOC113999927/ATXN1L/COL18A1/NF1/PDGFRA/SPARC/COL6A2 35 1035
GO:0035082 axoneme assembly 8/1035 21/12237 0.0001908 0.0456705 0.0449864 LOC113992699/LOC113990666/RSPH1/JHY/CFAP206/RSPH9/LOC113986306/LOC113984124 8 1035
GO:0018095 protein polyglutamylation 5/1035 10/12237 0.0007526 0.1601461 0.1577473 TTLL7/TPGS2/TTLL4/CEP41/TTLL11 5 1035
GO:0006811 ion transport 16/1035 81/12237 0.0010739 0.2056489 0.2025685 SLC22A15/STEAP1/SLC24A2/SLC26A1/LOC113988621/TRPA1/SLC12A2/LOC113991506/AQP11/WNK1/SLC26A6/GLRA1/LASP1/SLC1A3/LOC113989700/SLC24A3 16 1035



9 Nucleus Taenia (TnA)


9.1 Checking the sampling

For the TNA, we started with 16854 genes but after filering we have 14693 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have tna samples for 12, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=1
pilot run1 run2
floater 2 4 1
territorial 2 5 2
Sampling years distributed across batches, fisher test p=0.00075
2015 2017 2018
pilot 4 0 0
run1 0 6 3
run2 0 1 2


9.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"


9.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

PFT1 was previously identified as an outlier based on WGCNA dendrograms only. But here it’s not consistently dentified as an outlier.Also previous analysis shows that it doesn’t have a strong influence on the results, so I am going to keep it in this data. Looks like we only need to control for Batch, as it removes the effects of year on our data.


9.2 Status TNA

9.2.1 Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 14693 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 5, 0.034%
FALSE LFC < 0 (down)     : 8, 0.054%
FALSE outliers [1]       : 24, 0.16%
FALSE low counts [2]     : 1415, 9.6%
FALSE (mean count < 17)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


9.2.2 Gene Ontology

Top 10 enriched GO BP in TNA according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0010951 negative regulation of endopeptidase activity 16/963 70/11975 0.0001062 0.1979001 0.1978107 LOC113991485/ITIH5/ANXA2/LOC113992817/LOC113991484/SPINT1/APP/LOC113992168/CRIM1/SERPINB1/USP14/TIMP4/A2M/SERPINE1/LOC113991482/LOC113991487 16 963
GO:0060291 long-term synaptic potentiation 10/963 36/11975 0.0003965 0.2559777 0.2558621 SYT12/CRHR2/NLGN1/ARC/SLC24A2/PLK2/PTN/CRH/ITPR3/LRRTM2 10 963
GO:0051965 positive regulation of synapse assembly 12/963 50/11975 0.0004723 0.2559777 0.2558621 EPHB2/CUX2/AMIGO3/NLGN1/PTPRD/LRRN3/SLITRK6/EEF2K/SLITRK2/IL1RAP/LRRTM3/LRRTM2 12 963
GO:0006614 SRP-dependent cotranslational protein targeting to membrane 16/963 81/11975 0.0006227 0.2559777 0.2558621 LOC114002081/LOC113998898/RPL4/RPL31/RPL19/RPS27A/RPS23/RPL37/RPS21/RPL7/RPS29/RPL30/RPS5/RPL23A/SRP19/RPS3A 16 963
GO:0006936 muscle contraction 15/963 74/11975 0.0006866 0.2559777 0.2558621 ITGA1/CALD1/GAMT/VCL/TPM2/CACNA1H/SORBS3/ITGA11/LOC113993584/LOC113999652/PXN/CALM1/LOC113999542/LOC113999078/LOC114001935 15 963
GO:0007093 mitotic cell cycle checkpoint 5/963 12/11975 0.0016323 0.3612802 0.3611170 NABP2/CHFR/PLK2/PLK3/NABP1 5 963
GO:0090050 positive regulation of cell migration involved in sprouting angiogenesis 7/963 23/11975 0.0016572 0.3612802 0.3611170 RHOJ/PLK2/VEGFA/PTGS2/KDR/LOC113997171/LOC114000824 7 963
GO:0035633 maintenance of blood-brain barrier 8/963 30/11975 0.0020087 0.3612802 0.3611170 TJP3/ITGB1/LOC114001565/CDH5/VCL/CLDN1/TJP1/SLC12A2 8 963
GO:0042127 regulation of cell population proliferation 18/963 107/11975 0.0020783 0.3612802 0.3611170 LOC114000703/SIX3/TOPORS/LTK/LOC114000971/MEAK7/XIAP/DUSP15/MAFG/JUN/IL4R/TGFB3/LOC114001747/CCDC88A/SHH/SLC7A11/AGTR1/LOC114000824 18 963
GO:0090141 positive regulation of mitochondrial fission 6/963 18/11975 0.0021270 0.3612802 0.3611170 MIEF2/BNIP3/MIEF1/KDR/LOC113997171/SPIRE1 6 963


9.3 Testosterone TNA

9.3.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

9.3.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

Taking the batch effect into account seems to remove the association with Year too. Only need to include Batch in the model.

9.3.3 Analysis

Formula = ~ Batch + mean_T

FALSE 
FALSE out of 14614 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 99, 0.68%
FALSE LFC < 0 (down)     : 143, 0.98%
FALSE outliers [1]       : 37, 0.25%
FALSE low counts [2]     : 280, 1.9%
FALSE (mean count < 6)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


9.3.4 Gene Ontology

Top 10 enriched GO BP in TNA according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0060291 long-term synaptic potentiation 13/1251 36/11931 0.0000398 0.0798071 0.0794405 SLC24A2/CRHR2/PTN/SYT12/PLK2/NLGN1/NR2E1/RASGRF2/LOC114003174/RGS14/SLC1A1/LOC113985801/MECP2 13 1251
GO:0010951 negative regulation of endopeptidase activity 18/1251 69/11931 0.0001951 0.1574843 0.1567608 LOC113992168/ANXA2/SPOCK1/ITIH2/SPOCK3/TFPI2/PROS1/TIMP4/SPINT1/SERPIND1/LOC113999601/A2M/LOC113991484/LOC113992708/SERPINB5/APLP2/SERPINF1/APP 18 1251
GO:0045446 endothelial cell differentiation 6/1251 11/11931 0.0003808 0.1574843 0.1567608 PTN/JAG2/CXCR4/S1PR1/LOC113997171/SCUBE1 6 1251
GO:0001755 neural crest cell migration 13/1251 44/11931 0.0004024 0.1574843 0.1567608 RADIL/LOC113982696/LOC113984504/SEMA4G/RET/KITLG/SEMA4C/SEMA3C/LOC114000376/SEMA3A/SEMA3B/SEMA4B/SEMA4A 13 1251
GO:0048843 negative regulation of axon extension involved in axon guidance 9/1251 24/11931 0.0004508 0.1574843 0.1567608 LOC113982696/LOC113984504/SEMA4G/SEMA4C/SEMA3C/SEMA3A/SEMA3B/SEMA4B/SEMA4A 9 1251
GO:0071526 semaphorin-plexin signaling pathway 11/1251 34/11931 0.0004713 0.1574843 0.1567608 LOC113982696/LOC113984504/SEMA4G/SEMA4C/MET/SEMA3C/SEMA3A/SEMA3B/SEMA4B/LOC113986256/SEMA4A 11 1251
GO:0050919 negative chemotaxis 11/1251 38/11931 0.0013385 0.2642404 0.2630265 LOC113982696/LOC113984504/SEMA4G/SEMA4C/RHOA/SEMA3C/SEMA3A/SEMA3B/SEMA4B/FLRT3/SEMA4A 11 1251
GO:0007613 memory 17/1251 74/11931 0.0013965 0.2642404 0.2630265 SLC24A2/CEBPB/PPP1R1B/PTN/LGMN/FGF13/LOC113991532/PLK2/KALRN/CCK/ASIC3/KCNK2/LOC113989303/LOC114003174/ADCY8/LOC113985801/PAK5 17 1251
GO:1990573 potassium ion import across plasma membrane 10/1251 33/11931 0.0014975 0.2642404 0.2630265 KCNJ6/SLC12A2/LOC113999542/FXYD2/LOC113996443/ATP1B1/LOC113986621/KCNJ3/ATP1B3/ATP1A1 10 1251
GO:0070374 positive regulation of ERK1 and ERK2 cascade 26/1251 136/11931 0.0017453 0.2642404 0.2630265 CALCR/FGFR4/CFLAR/LOC113991532/FGFR3/RAP1B/GPER1/CX3CL1/PDGFC/LOC113997171/MAP2K1/APELA/PRKCA/SRC/HTR2C/FGFR2/BMP4/ANGPT4/ALKAL2/P2RY1/RIPK2/BMP2/DDT/PTPRC/SLC30A10/APP 26 1251


9.4 mean T x Status

FALSE 
FALSE out of 14614 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 29, 0.2%
FALSE LFC < 0 (down)     : 13, 0.089%
FALSE outliers [1]       : 22, 0.15%
FALSE low counts [2]     : 3947, 27%
FALSE (mean count < 94)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


9.5 Strength TNA

9.5.1 Analysis

Formula ~ Batch + scale(Strength)

FALSE 
FALSE out of 14614 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 22, 0.15%
FALSE LFC < 0 (down)     : 34, 0.23%
FALSE outliers [1]       : 38, 0.26%
FALSE low counts [2]     : 839, 5.7%
FALSE (mean count < 11)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


9.5.2 Gene Ontology

Top 10 enriched GO BP in TNA according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0050771 negative regulation of axonogenesis 7/911 20/11931 0.0004733 0.4902006 0.4902006 LOC113988556/DAB1/LOC113990764/LRP4/RUFY3/LINGO1/RTN4 7 911
GO:0043951 negative regulation of cAMP-mediated signaling 6/911 15/11931 0.0005352 0.4902006 0.4902006 LOC113994469/LOC113993433/LOC113993299/LPAR1/APLNR/PDE3A 6 911
GO:2000310 regulation of NMDA receptor activity 8/911 31/11931 0.0018130 0.7677461 0.7677461 MEF2C/DAPK1/ACTN2/DLG3/OPRM1/GRIA4/GRIA1/RASGRF2 8 911
GO:0034975 protein folding in endoplasmic reticulum 5/911 13/11931 0.0019666 0.7677461 0.7677461 HSPA5/HSP90B1/PDIA3/CALR/DNAJC3 5 911
GO:0035023 regulation of Rho protein signal transduction 7/911 25/11931 0.0020954 0.7677461 0.7677461 ARHGEF28/ARHGDIB/ARHGEF10/STARD8/LOC114000754/AKAP13/RASGRF2 7 911
GO:0007186 G protein-coupled receptor signaling pathway 50/911 446/11931 0.0037799 0.8309344 0.8309344 LOC113994469/HTR1D/LANCL1/PLEKHG5/LOC113993090/GNAI3/PREX1/CCKBR/LOC113994699/LOC113993433/LOC113983719/CX3CR1/P2RY13/LOC113989258/ARHGEF26/ADORA1/LOC113993299/OPRM1/PLXNB1/ABR/GPRC5C/LPAR1/APLNR/TACR1/LPAR2/AKAP13/GRP/RIC8B/GPR85/GRM2/GRM7/RASGRF2/GPR162/GPR146/CARTPT/FZD5/QRFPR/PDE3A/AVPR1A/CALCR/LOC113983132/DGKQ/ARHGEF11/RAMP3/HTR1E/ADGRG2/CCK/CRHR2/LOC114003560/POMC 50 911
GO:0006986 response to unfolded protein 8/911 35/11931 0.0041211 0.8309344 0.8309344 FICD/HSP90AB1/HSPA8/HSP90AA1/MANF/UBE2J2/HSPA4/ERP27 8 911
GO:0070166 enamel mineralization 4/911 10/11931 0.0048811 0.8309344 0.8309344 TCIRG1/FAM20A/LOC113984874/STIM1 4 911
GO:0086004 regulation of cardiac muscle cell contraction 4/911 10/11931 0.0048811 0.8309344 0.8309344 ATP1A1/KCNJ2/ADORA1/LOC113993299 4 911
GO:0030217 T cell differentiation 6/911 22/11931 0.0050256 0.8309344 0.8309344 TCIRG1/DLL4/PREX1/PTPRC/PATZ1/IL7R 6 911



10 Arcopallium intermedium (Ai)


10.1 Checking the sampling

For the AI, we started with 16854 genes but after filering we have 14598 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have ai samples for 11, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=0.54
pilot run1
floater 2 4
territorial 1 7
Sampling years distributed across batches, fisher test p=0.0027
2015 2017 2018
pilot 3 0 0
run1 0 6 5


10.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE null device 
FALSE           1


10.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

Removing the batch effect effectively controls for Year.


10.2 Status AI

10.2.1 Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 14598 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 11, 0.075%
FALSE LFC < 0 (down)     : 7, 0.048%
FALSE outliers [1]       : 10, 0.069%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 1)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


10.2.2 Gene Ontology

Top 10 enriched GO BP in AI according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0071470 cellular response to osmotic stress 5/463 13/11914 0.0000862 0.1158070 0.1158070 LOC113991485/LOC113991484/LOC113991487/LOC113991482/LOC114001824 5 463
GO:0002088 lens development in camera-type eye 6/463 27/11914 0.0004919 0.2676867 0.2676867 GJE1/DLG1/PITX3/PROX1/UNC45A/CRYBB2 6 463
GO:0045670 regulation of osteoclast differentiation 4/463 11/11914 0.0005975 0.2676867 0.2676867 BGLAP/CEBPB/CAMK4/MITF 4 463
GO:0006271 DNA strand elongation involved in DNA replication 4/463 12/11914 0.0008689 0.2919463 0.2919463 GINS1/MCM4/POLD2/POLD3 4 463
GO:0043278 response to morphine 5/463 23/11914 0.0016327 0.4215656 0.4215656 SRR/PENK/CXCR4/PITX3/LOC113984425 5 463
GO:0032201 telomere maintenance via semi-conservative replication 5/463 25/11914 0.0024181 0.4215656 0.4215656 POLE2/POLE3/LOC114000574/POLD2/POLD3 5 463
GO:0035023 regulation of Rho protein signal transduction 5/463 25/11914 0.0024181 0.4215656 0.4215656 LOC114003556/ARHGEF10/ARHGDIB/STARD8/PDCD10 5 463
GO:0006260 DNA replication 12/463 121/11914 0.0025939 0.4215656 0.4215656 POLE2/POLE3/LOC114000574/MCM4/SDE2/MCM5/BRCA1/CDK2/NFIA/ORC6/LOC114001320/POLD2 12 463
GO:0001676 long-chain fatty acid metabolic process 4/463 16/11914 0.0028230 0.4215656 0.4215656 ACAD9/ACSL4/ACSL1/SLC27A6 4 463
GO:0010951 negative regulation of endopeptidase activity 8/463 67/11914 0.0042717 0.5626977 0.5626977 LOC113991485/LOC113992168/LOC113991484/LOC113991487/LOC113992709/LOC113991482/C3/CRB2 8 463


10.3 Testosterone AI

10.3.1 Expression data exploration

FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

No association with year.


10.3.2 Analysis

Formula = ~ mean_T

FALSE 
FALSE out of 14628 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 13, 0.089%
FALSE LFC < 0 (down)     : 30, 0.21%
FALSE outliers [1]       : 25, 0.17%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


10.3.3 Gene Ontology

Top 10 enriched GO BP in AI according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0060438 trachea development 5/490 10/11926 0.0000244 0.0273859 0.0273859 LOC113987257/LOC113994177/LOC114000433/LOC113999122/LOC113992388 5 490
GO:0002064 epithelial cell development 6/490 17/11926 0.0000392 0.0273859 0.0273859 LOC113987257/LOC113994177/LOC114000433/LOC113999122/LOC113992388/CXCR4 6 490
GO:1904158 axonemal central apparatus assembly 5/490 12/11926 0.0000715 0.0332832 0.0332832 LOC113987257/LOC113994177/LOC114000433/LOC113999122/LOC113992388 5 490
GO:0010976 positive regulation of neuron projection development 13/490 107/11926 0.0004303 0.1502731 0.1502731 PTN/CNTN1/NDRG4/LOC113990140/CAMK1/LOC113996443/CFLAR/TWF1/AMIGO1/LOC114002148/FYN/PLPPR5/RGS2 13 490
GO:0072657 protein localization to membrane 5/490 18/11926 0.0006299 0.1759910 0.1759910 TM9SF3/TM9SF2/HPS6/HGS/PACSIN1 5 490
GO:0016082 synaptic vesicle priming 5/490 20/11926 0.0010648 0.2479230 0.2479230 LOC113991140/SNAP29/STXBP1/LOC113993758/LOC113987015 5 490
GO:0061025 membrane fusion 7/490 42/11926 0.0014639 0.2623042 0.2623042 LOC113991140/SNAP29/LOC113993758/DNM1L/LOC113987015/LOC113993578/DNM3 7 490
GO:0021591 ventricular system development 5/490 22/11926 0.0016899 0.2623042 0.2623042 LOC113987257/LOC113994177/LOC114000433/LOC113999122/LOC113992388 5 490
GO:0045742 positive regulation of epidermal growth factor receptor signaling pathway 5/490 22/11926 0.0016899 0.2623042 0.2623042 AFAP1L2/PDE6G/LOC113992190/HAP1/CNOT9 5 490
GO:0006749 glutathione metabolic process 6/490 36/11926 0.0031798 0.4386326 0.4386326 LOC113990807/LOC113990865/LOC113989543/SLC1A1/LOC113984987/LOC113990605 6 490


10.4 mean T x Status

FALSE 
FALSE out of 14628 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 4, 0.027%
FALSE LFC < 0 (down)     : 3, 0.021%
FALSE outliers [1]       : 493, 3.4%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


10.5 Strength AI

10.5.1 Analysis

Formula ~ scale(Strength)

FALSE 
FALSE out of 14628 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 18, 0.12%
FALSE LFC < 0 (down)     : 16, 0.11%
FALSE outliers [1]       : 21, 0.14%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


10.5.2 Gene Ontology

Top 10 enriched GO BP in AI according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0055072 iron ion homeostasis 5/274 24/11926 0.0001833 0.1303095 0.1290652 B2M/HEPHL1/TMPRSS6/PICALM/STEAP3 5 274
GO:0030198 extracellular matrix organization 14/274 205/11926 0.0002749 0.1303095 0.1290652 HSPG2/LOC113999927/ADAMTS14/SERPINB5/VWA1/COL4A4/COL1A1/ITGA2B/COL7A1/CSGALNACT1/ITGA4/ITGB6/MMP23B/LOC113989166 14 274
GO:0043122 regulation of I-kappaB kinase/NF-kappaB signaling 4/274 24/11926 0.0020153 0.5075303 0.5026838 UFL1/ZFAND6/SQSTM1/ANGPT1 4 274
GO:0045765 regulation of angiogenesis 4/274 27/11926 0.0031535 0.5075303 0.5026838 EGLN1/HTATIP2/ADGRA2/BMPER 4 274
GO:0007611 learning or memory 5/274 46/11926 0.0039183 0.5075303 0.5026838 B2M/VIP/SLC1A1/CRHBP/PICALM 5 274
GO:0019222 regulation of metabolic process 3/274 15/11926 0.0044469 0.5075303 0.5026838 MC3R/LOC113983719/MC5R 3 274
GO:0051289 protein homotetramerization 5/274 48/11926 0.0047163 0.5075303 0.5026838 CBY1/B2M/LOC113984533/CBR4/KIF25 5 274
GO:0043588 skin development 4/274 31/11926 0.0052622 0.5075303 0.5026838 DACT2/COL1A1/LOC113984533/ITGB6 4 274
GO:2001224 positive regulation of neuron migration 3/274 16/11926 0.0053811 0.5075303 0.5026838 SEMA6A/LOC114003556/SEMA3A 3 274
GO:0044267 cellular protein metabolic process 9/274 144/11926 0.0059933 0.5075303 0.5026838 HSPG2/VWA1/ALB/B2M/CDH2/TSPAN5/MFGE8/APH1A/IGF2 9 274



11 Lateral Septum (Ls)


11.1 Checking the sampling

For the LS, we started with 16854 genes but after filering we have 14899 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have ls samples for 12, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=1
pilot run2
floater 1 5
territorial 2 7
Sampling years distributed across batches, fisher test p=0.0022
2015 2017 2018
pilot 3 0 0
run2 0 7 5

11.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

11.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

PFT2 was previously identified as an outlier, but using the quantitative threshold and controlling for batch it doesn’t come out as different. Controlling for Batch adequately controls for Year.

11.2 Status LS

11.2.1 Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 14899 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 2, 0.013%
FALSE LFC < 0 (down)     : 3, 0.02%
FALSE outliers [1]       : 16, 0.11%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 1)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


11.2.2 Gene Ontology

Top 10 enriched GO BP in LS according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0036159 inner dynein arm assembly 7/609 15/12088 0.0000036 0.0057326 0.0056371 LRRC6/LOC113981955/DNAAF5/CFAP73/DNAH1/DNAAF1/DNAH12 7 609
GO:0003341 cilium movement 10/609 38/12088 0.0000128 0.0102328 0.0100622 ARMC4/LRRC6/LOC113981955/DNAAF5/CFAP73/DNAH1/TTC29/DNAAF1/DNAH12/DNAH9 10 609
GO:0071377 cellular response to glucagon stimulus 7/609 22/12088 0.0000699 0.0371562 0.0365370 LOC113984696/CPS1/LOC113995499/ADCY5/PRKAR2A/LOC113999236/LOC113983744 7 609
GO:0060271 cilium assembly 22/609 181/12088 0.0001126 0.0416889 0.0409942 TMEM237/CC2D2A/IFT88/WDR35/CEP126/CFAP54/IFT74/TCTN2/LOC113992699/ARL13B/INTU/CEP89/NUDCD3/CFAP44/POC1B/ATXN10/KIF27/DNAAF1/IQCB1/SSX2IP/TMEM67/PCNT 22 609
GO:0007018 microtubule-based movement 12/609 68/12088 0.0001307 0.0416889 0.0409942 LOC114000985/DNAH10/LOC113982299/DNAH3/KIFC3/DNAH1/KIF6/KIF27/KIF21B/DNAH12/DNAH9/LOC113996658 12 609
GO:0007190 activation of adenylate cyclase activity 8/609 33/12088 0.0001794 0.0476815 0.0468869 LOC113984696/CALCR/DRD5/LOC114001840/ADCY5/ADORA3/LOC113999236/LOC113983744 8 609
GO:0030317 flagellated sperm motility 8/609 35/12088 0.0002781 0.0633726 0.0623166 CFAP65/LRRC6/TCTE1/DNAH1/PLTP/DNAJA1/LOC113999236/LOC113983744 8 609
GO:0015718 monocarboxylic acid transport 5/609 14/12088 0.0004370 0.0712075 0.0700209 SLC16A9/SLC16A5/LOC113990781/LOC113996953/SLC6A12 5 609
GO:0060285 cilium-dependent cell motility 5/609 14/12088 0.0004370 0.0712075 0.0700209 DNAH3/CFAP44/DNAH1/DNAH12/EFHC2 5 609
GO:0007283 spermatogenesis 24/609 226/12088 0.0004464 0.0712075 0.0700209 LOC113982299/SPEF2/GJA1/GAMT/MSH4/CFAP54/LOC113992699/NUP62CL/SEPT4/ACVR2A/DHX36/LOC113987746/KATNAL1/LOC113992196/MNS1/DNMT3A/TDRD9/DNAJA1/UBR2/GGT7/PUM1/STK11/LOC113985256/LOC113996658 24 609


11.3 Testosterone LS

11.3.1 Expression data exploration

FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE


11.3.2 Analysis

Formula = ~ mean_T

FALSE 
FALSE out of 14818 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 0, 0%
FALSE LFC < 0 (down)     : 2, 0.013%
FALSE outliers [1]       : 62, 0.42%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 4)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


11.3.3 Gene Ontology

Top 10 enriched GO BP in LS according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0043525 positive regulation of neuron apoptotic process 7/385 45/12041 0.0005140 0.3788809 0.3646207 FASLG/BAX/LOC113991532/ITGA11/ATF4/PRNP/CDC42 7 385
GO:0022010 central nervous system myelination 4/385 13/12041 0.0005849 0.3788809 0.3646207 LOC113990764/ASPA/LOC113991947/PLP1 4 385
GO:0016525 negative regulation of angiogenesis 9/385 78/12041 0.0008260 0.3788809 0.3646207 CNMD/SPRY2/FASLG/THBS4/CLDN5/PLK2/APOH/SPARC/PDE3B 9 385
GO:0008217 regulation of blood pressure 7/385 54/12041 0.0015680 0.4929332 0.4743803 SGK1/STK39/POMC/SOD1/ACVRL1/CD34/LOC113992470 7 385
GO:0003151 outflow tract morphogenesis 6/385 41/12041 0.0017912 0.4929332 0.4743803 TBX2/NIPBL/SOX11/CLDN5/TGFBR3/HIF1A 6 385
GO:0010466 negative regulation of peptidase activity 3/385 10/12041 0.0032917 0.5361629 0.5159830 LOC113992709/LOC113992708/LOC113999601 3 385
GO:1900264 positive regulation of DNA-directed DNA polymerase activity 3/385 10/12041 0.0032917 0.5361629 0.5159830 PCNA/RFC4/RFC3 3 385
GO:0010738 regulation of protein kinase A signaling 3/385 11/12041 0.0044195 0.5361629 0.5159830 AIP/PRKAR1A/AKAP12 3 385
GO:0070987 error-free translesion synthesis 4/385 22/12041 0.0047659 0.5361629 0.5159830 PCNA/RFC4/RCHY1/RFC3 4 385
GO:0045778 positive regulation of ossification 3/385 12/12041 0.0057542 0.5361629 0.5159830 NIPBL/SOX11/WNT5A 3 385


11.4 mean T x Status

FALSE 
FALSE out of 14818 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 16, 0.11%
FALSE LFC < 0 (down)     : 23, 0.16%
FALSE outliers [1]       : 29, 0.2%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 4)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


11.5 Strength LS

11.5.1 Analysis

Formula ~ scale(Strength)

FALSE 
FALSE out of 14818 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 17, 0.11%
FALSE LFC < 0 (down)     : 81, 0.55%
FALSE outliers [1]       : 53, 0.36%
FALSE low counts [2]     : 6849, 46%
FALSE (mean count < 303)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


11.5.2 Gene Ontology

Top 10 enriched GO BP in LS according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0045176 apical protein localization 8/1426 14/12041 0.0000589 0.1214821 0.1214821 SHROOM2/ARF4/GOPC/RDX/ATP8B1/INSC/JAM3/SHROOM3 8 1426
GO:0045116 protein neddylation 6/1426 11/12041 0.0007430 0.6108154 0.6108154 DCUN1D1/UBA3/DCUN1D4/RNF7/UBE2M/DCUN1D2 6 1426
GO:0006888 endoplasmic reticulum to Golgi vesicle-mediated transport 32/1426 153/12041 0.0008891 0.6108154 0.6108154 COG1/RAB1A/USE1/GORASP1/COL7A1/MCFD2/TMED2/TMED7/ARF4/RAB2A/DYNC1I1/TMED5/TRAPPC2/DYNC1LI1/CNIH1/GOPC/KDELR2/TMED4/LMAN1/CAPZA2/SPTBN1/COG6/ANK2/LMAN2L/SEC24C/SPAST/COG4/SCFD1/SEC24B/VCP/BET1/TEX261 32 1426
GO:0006904 vesicle docking involved in exocytosis 11/1426 34/12041 0.0013235 0.6213255 0.6213255 RAB12/STXBP1/VPS45/RAB3A/LOC113982219/VPS33B/VPS11/SCFD1/RAB8B/LOC113997864/LOC114001550 11 1426
GO:0015031 protein transport 53/1426 298/12041 0.0015073 0.6213255 0.6213255 COG1/GPM6B/USE1/PSEN1/SCAMP1/GORASP1/GOLT1B/MCFD2/SNAP29/MLC1/LIN7C/ATG4D/SNX3/SNX24/RAB2A/RAB9A/RAB39A/GABARAPL2/GOPC/SFT2D1/APBA2/RAB33B/LOC113982219/RP2/KDELR2/PTPN23/VPS33B/ATG4C/YIPF5/WASHC3/PIK3R2/LOC113995129/COG6/ANK2/ARRDC4/LMAN2L/VTA1/COG4/EXOC2/VPS52/PHAX/RUFY1/BET1/COL1A1/LOC113992565/LOC113996864/LOC113997864/LOC114000510/LOC114001550/MYO1D/RAMP1/S100A13/TESC 53 1426
GO:0050821 protein stabilization 32/1426 160/12041 0.0019270 0.6619383 0.6619383 LAMP1/STXBP1/SUGT1/BAG1/LAMP2/CAMLG/RAB21/CHP1/NOP53/NAA16/WFS1/NLK/EP300/CCNH/GNAQ/ANK2/ATP1B1/HSP90AB1/STX12/USP36/SUMO1/BAG4/HYPK/USP7/PFN2/LOC113990352/PTEN/CRTAP/LOC113992637/TAF9B/TESC/UBE2B 32 1426
GO:0043405 regulation of MAP kinase activity 5/1426 10/12041 0.0034880 0.7546298 0.7546298 KSR1/DUSP19/HGS/TRIB2/PPEF2 5 1426
GO:0097502 mannosylation 5/1426 10/12041 0.0034880 0.7546298 0.7546298 ALG12/ALG1/ALG9/ALG2/ALG3 5 1426
GO:0008654 phospholipid biosynthetic process 8/1426 23/12041 0.0036128 0.7546298 0.7546298 LOC113988212/MBOAT2/AGPAT5/GPAM/FITM2/LPGAT1/PCYT2/LOC113992314 8 1426
GO:0070979 protein K11-linked ubiquitination 9/1426 28/12041 0.0037776 0.7546298 0.7546298 UBE2D3/UBE2W/UBE2E3/UBE2A/FZR1/ANAPC1/RNF4/UBE2L3/UBE2B 9 1426



12 Bed Nucleus of the Stria Terminalis (BSTm)


12.1 Checking the sampling

For the BSTm, we started with 16854 genes but after filering we have 15031 genes. Before I go into more detail, I am going to check that the sampling is equal across batches with respect to our variables of interest. With respect to Status we have bstm samples for 12, with respect to T and the tracking data.

Status across sequencing runs, fisher test p=1
pilot run2
floater 2 5
territorial 2 7
Sampling years distributed across batches, fisher test p=0.00055
2015 2017 2018
pilot 4 0 0
run2 0 7 5


12.1.1 Expression data exploration

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"


12.1.2 Corrections applied

FALSE [1] "Batch"
FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE

Batch controls effectively for Year.


12.2 Status BSTm

12.2.1 Analysis

Formula = ~ Batch + Status

FALSE 
FALSE out of 15031 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 6, 0.04%
FALSE LFC < 0 (down)     : 3, 0.02%
FALSE outliers [1]       : 5, 0.033%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


12.2.2 Gene Ontology

Top 10 enriched GO BP in BSTm according to Status
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0048706 embryonic skeletal system development 8/1014 21/12177 0.0001711 0.2413900 0.2411523 COL1A1/SP3/WNT11/NOG/PBX1/SHH/LOC113989878/MBTD1 8 1014
GO:0061564 axon development 7/1014 17/12177 0.0002509 0.2413900 0.2411523 DYNLT1/NEFH/LOC114000515/PLP1/LOC113996738/LOC113991704/TSPAN2 7 1014
GO:0001666 response to hypoxia 23/1014 136/12177 0.0008088 0.5010804 0.5005869 CHRNA4/VEGFC/BECN1/BNIP3/ITPR2/LOC113992301/SMAD4/CD24/DPP4/LOC113989812/SOD2/THBS1/HSD11B2/UCP3/WDR83/PRKAA1/PKM/PLOD1/P2RX3/CAT/SRF/HIF1A/ADORA1 23 1014
GO:0021762 substantia nigra development 10/1014 39/12177 0.0010426 0.5010804 0.5005869 CKB/LOC114001565/RAD1/ZNF148/GNB4/PLP1/ATP5PF/CNP/GLUD1/LOC114000755 10 1014
GO:0006091 generation of precursor metabolites and energy 9/1014 34/12177 0.0014454 0.5010804 0.5005869 SLC25A4/XYLB/ENPP1/LOC113996683/MCHR1/LOC113983432/PHKA2/ACO2/LOC113986170 9 1014
GO:0051028 mRNA transport 9/1014 35/12177 0.0018042 0.5010804 0.5005869 XPO1/IWS1/LOC113992455/ZC3H3/LOC113992247/IGF2BP2/IGF2BP3/CETN3/NUTF2 9 1014
GO:0021522 spinal cord motor neuron differentiation 6/1014 17/12177 0.0018231 0.5010804 0.5005869 ISL1/SOX4/PTCH1/SHH/LOC113997538/OLIG2 6 1014
GO:0001501 skeletal system development 17/1014 96/12177 0.0022457 0.5400944 0.5395625 HAPLN2/LOC113993027/COL1A1/MMP16/TLL1/CNMD/SOX4/UFD1/NOG/ACAN/CHRD/PAPSS2/USP1/HAPLN4/IGF1/HDAC4/TNFRSF11B 17 1014
GO:0009653 anatomical structure morphogenesis 14/1014 74/12177 0.0028417 0.6050786 0.6044827 DKK3/SOX4/SMAD4/SIAH1/FOXD1/BICD1/SMAD7/CPM/SOX21/TRIM13/IGF2BP2/SMAD5/F2R/IGF2BP3 14 1014
GO:0006120 mitochondrial electron transport, NADH to ubiquinone 9/1014 38/12177 0.0033228 0.6050786 0.6044827 NDUFB3/NDUFB5/NDUFV3/NDUFA4/NDUFS1/COQ9/NDUFB8/NDUFA12/NDUFA6 9 1014


12.3 Testosterone BSTm

12.3.1 Expression data exploration

FALSE [1] "Status"
FALSE [1] "Year"

FALSE  Flagging genes and samples with too many missing values...
FALSE   ..step 1
FALSE [1] TRUE


12.3.2 Analysis

Formula = ~ mean_T

FALSE 
FALSE out of 15010 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 0, 0%
FALSE LFC < 0 (down)     : 3, 0.02%
FALSE outliers [1]       : 32, 0.21%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


12.3.3 Gene Ontology

Top 10 enriched GO BP in BSTm according to mean_T
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:0055119 relaxation of cardiac muscle 4/585 11/12171 0.0013297 0.866266 0.866266 ATP1B1/ATP2A2/GSN/KCNJ2 4 585
GO:0021591 ventricular system development 5/585 19/12171 0.0016710 0.866266 0.866266 LOC114000433/LOC113997903/LOC113994177/NME5/SEMA6D 5 585
GO:0034614 cellular response to reactive oxygen species 6/585 29/12171 0.0022134 0.866266 0.866266 PRDX3/MAPK1/MPV17/MAPK9/MMP2/PDGFRA 6 585
GO:0000729 DNA double-strand break processing 4/585 15/12171 0.0047171 0.866266 0.866266 LOC113998423/NBN/SMARCAD1/DNA2 4 585
GO:0006695 cholesterol biosynthetic process 6/585 34/12171 0.0051131 0.866266 0.866266 GGPS1/MSMO1/LOC113997963/DHCR24/LOC114002832/PLPP6 6 585
GO:0019233 sensory perception of pain 6/585 34/12171 0.0051131 0.866266 0.866266 NPY1R/MMEL1/MAPK1/KCND2/NIPSNAP1/PRDM12 6 585
GO:0051932 synaptic transmission, GABAergic 4/585 16/12171 0.0060536 0.866266 0.866266 GABRG1/RAC3/GABRA2/LOC113990089 4 585
GO:0031623 receptor internalization 6/585 38/12171 0.0089263 0.866266 0.866266 ACKR3/CALCRL/CAV1/CD36/CXCR2/LOC113993578 6 585
GO:0006891 intra-Golgi vesicle-mediated transport 5/585 28/12171 0.0098851 0.866266 0.866266 RAB33B/COG5/COG6/RAB6A/VTI1A 5 585
GO:0006693 prostaglandin metabolic process 3/585 10/12171 0.0102877 0.866266 0.866266 LOC114004119/PTGR2/ZADH2 3 585


12.4 mean T x Status

FALSE 
FALSE out of 15010 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 3, 0.02%
FALSE LFC < 0 (down)     : 2, 0.013%
FALSE outliers [1]       : 11, 0.073%
FALSE low counts [2]     : 0, 0%
FALSE (mean count < 5)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


12.5 Strength BSTm

12.5.1 Analysis

Formula ~ scale(Strength)

FALSE 
FALSE out of 15010 with nonzero total read count
FALSE adjusted p-value < 0.1
FALSE LFC > 0 (up)       : 8, 0.053%
FALSE LFC < 0 (down)     : 7, 0.047%
FALSE outliers [1]       : 26, 0.17%
FALSE low counts [2]     : 870, 5.8%
FALSE (mean count < 11)
FALSE [1] see 'cooksCutoff' argument of ?results
FALSE [2] see 'independentFiltering' argument of ?results

FALSE null device 
FALSE           1


12.5.2 Gene Ontology

Top 10 enriched GO BP in BSTm according to Strength
Description GeneRatio BgRatio pvalue p.adjust qvalue geneID Count n_annotated
GO:1903077 negative regulation of protein localization to plasma membrane 7/979 14/12171 0.0000442 0.0853381 0.0846167 PICALM/PID1/MRAP2/BCL2L1/TMEM59/PPFIA1/TMBIM1 7 979
GO:0072659 protein localization to plasma membrane 24/979 141/12171 0.0003458 0.1840597 0.1825037 LOC113988212/GPR158/LOC114002160/RAB3C/SLC4A1/KCNIP4/CDH2/ATP1B1/DPP6/MRAP2/LOC113988625/RAB15/PKP2/IFT20/ATP1B3/ZDHHC3/DLG1/EMP2/RAB12/ACTN2/TPBG/EPB41L3/TTC7B/GAK 24 979
GO:0030225 macrophage differentiation 6/979 14/12171 0.0004558 0.1840597 0.1825037 MMP9/VEGFA/CSF1R/CDC42/SIRT1/BMP4 6 979
GO:0043647 inositol phosphate metabolic process 11/979 44/12171 0.0005573 0.1840597 0.1825037 PLCD3/PLCH2/INPP5J/IPMK/PLCB4/PLCG1/PTEN/PLCG2/INPPL1/MINPP1/PPIP5K2 11 979
GO:0043588 skin development 9/979 32/12171 0.0006996 0.1840597 0.1825037 PKD1/DACT2/PDGFA/LOC113984010/SLITRK5/COL5A1/LOC113998649/PTCH2/FRAS1 9 979
GO:0030534 adult behavior 8/979 26/12171 0.0007167 0.1840597 0.1825037 PCDH17/LOC113984666/LOC113995009/GABRG2/CNTNAP2/PTEN/SLITRK5/LOC113982203 8 979
GO:0042391 regulation of membrane potential 16/979 82/12171 0.0007195 0.1840597 0.1825037 GABRG1/GABRA5/PXK/CHRNB3/GLRA2/GABRG2/KCNMA1/LOC113989700/LOC113999820/CHRNA7/GLRB/HCN1/DLG1/GABRA2/ACTN2/LOC113996568 16 979
GO:0007268 chemical synaptic transmission 30/979 202/12171 0.0007629 0.1840597 0.1825037 GABRG1/GABRA5/NPTX2/GALR3/SV2C/CHRNB3/LOC113984666/GLRA2/LOC113995009/GABRG2/LOC113992038/LOC113989700/LOC113999820/GAD2/CHRNA7/GLRB/APBA1/SLITRK5/KCND2/ATXN3/HTR7/DLG1/GABRA2/LOC114001565/PAFAH1B1/MPZL1/SLC6A1/SLC1A1/DTNA/LOC113996568 30 979
GO:0050877 nervous system process 10/979 40/12171 0.0009868 0.2116229 0.2098339 GABRG1/GABRA5/CHRNB3/GLRA2/GABRG2/LOC113999820/CHRNA7/GLRB/GABRA2/LOC113996568 10 979
GO:0032228 regulation of synaptic transmission, GABAergic 5/979 13/12171 0.0024811 0.4353275 0.4316474 KRAS/PLCL2/SYN3/PTEN/LOC113995955 5 979

Volcano plots and similarity of gene expression between tissues is in another Rmarkdown on this site.

FALSE R version 4.0.2 (2020-06-22)
FALSE Platform: x86_64-w64-mingw32/x64 (64-bit)
FALSE Running under: Windows 10 x64 (build 19045)
FALSE 
FALSE Matrix products: default
FALSE 
FALSE locale:
FALSE [1] LC_COLLATE=English_Australia.1252  LC_CTYPE=English_Australia.1252   
FALSE [3] LC_MONETARY=English_Australia.1252 LC_NUMERIC=C                      
FALSE [5] LC_TIME=English_Australia.1252    
FALSE 
FALSE attached base packages:
FALSE  [1] grid      parallel  stats4    stats     graphics  grDevices utils    
FALSE  [8] datasets  methods   base     
FALSE 
FALSE other attached packages:
FALSE  [1] dplyr_1.0.7                 ggnetwork_0.5.8            
FALSE  [3] igraph_1.2.6                ggfortify_0.4.11           
FALSE  [5] GGally_2.1.0                effectsize_0.8.6           
FALSE  [7] car_3.0-8                   carData_3.0-4              
FALSE  [9] viridis_0.6.1               viridisLite_0.4.0          
FALSE [11] rrvgo_1.0.2                 UpSetR_1.4.0               
FALSE [13] limma_3.44.3                flashClust_1.01-2          
FALSE [15] org.Hs.eg.db_3.11.4         AnnotationDbi_1.50.0       
FALSE [17] clusterProfiler_3.16.1      fdrtool_1.2.15             
FALSE [19] ggpubr_0.4.0                gridExtra_2.3              
FALSE [21] wesanderson_0.3.6           VennDiagram_1.6.20         
FALSE [23] futile.logger_1.4.3         kableExtra_1.3.1           
FALSE [25] WGCNA_1.69                  fastcluster_1.1.25         
FALSE [27] dynamicTreeCut_1.63-1       pheatmap_1.0.12            
FALSE [29] lubridate_1.8.0             DESeq2_1.28.1              
FALSE [31] SummarizedExperiment_1.18.1 DelayedArray_0.14.0        
FALSE [33] matrixStats_0.56.0          Biobase_2.48.0             
FALSE [35] GenomicRanges_1.40.0        GenomeInfoDb_1.24.0        
FALSE [37] IRanges_2.22.2              S4Vectors_0.26.1           
FALSE [39] BiocGenerics_0.34.0         stringr_1.4.0              
FALSE [41] PCAtools_2.0.0              cowplot_1.0.0              
FALSE [43] lattice_0.20-41             ggrepel_0.8.2              
FALSE [45] ggplot2_3.4.4               reshape2_1.4.4             
FALSE [47] plyr_1.8.6                 
FALSE 
FALSE loaded via a namespace (and not attached):
FALSE   [1] ragg_1.2.1                tidyr_1.1.4              
FALSE   [3] acepack_1.4.1             bit64_0.9-7              
FALSE   [5] knitr_1.29                irlba_2.3.3              
FALSE   [7] data.table_1.13.0         rpart_4.1-15             
FALSE   [9] RCurl_1.98-1.2            doParallel_1.0.15        
FALSE  [11] generics_0.0.2            snow_0.4-3               
FALSE  [13] preprocessCore_1.50.0     lambda.r_1.2.4           
FALSE  [15] RSQLite_2.2.0             europepmc_0.4            
FALSE  [17] bit_4.0.4                 enrichplot_1.8.1         
FALSE  [19] webshot_0.5.2             xml2_1.3.2               
FALSE  [21] httpuv_1.5.4              assertthat_0.2.1         
FALSE  [23] xfun_0.29                 hms_1.1.1                
FALSE  [25] jquerylib_0.1.4           evaluate_0.14            
FALSE  [27] promises_1.1.1            fansi_0.4.1              
FALSE  [29] progress_1.2.2            readxl_1.3.1             
FALSE  [31] DBI_1.1.0                 geneplotter_1.66.0       
FALSE  [33] htmlwidgets_1.5.3         reshape_0.8.8            
FALSE  [35] purrr_0.3.4               ellipsis_0.3.2           
FALSE  [37] backports_1.1.7           insight_0.19.6           
FALSE  [39] annotate_1.66.0           gridBase_0.4-7           
FALSE  [41] vctrs_0.6.5               abind_1.4-5              
FALSE  [43] withr_2.5.2               ggforce_0.3.2            
FALSE  [45] triebeard_0.3.0           treemap_2.4-2            
FALSE  [47] checkmate_2.0.0           prettyunits_1.1.1        
FALSE  [49] cluster_2.1.0             DOSE_3.14.0              
FALSE  [51] crayon_1.4.2              genefilter_1.70.0        
FALSE  [53] pkgconfig_2.0.3           slam_0.1-47              
FALSE  [55] labeling_0.3              tweenr_1.0.1             
FALSE  [57] nlme_3.1-148              wordcloud_2.6            
FALSE  [59] nnet_7.3-14               rlang_1.1.2              
FALSE  [61] lifecycle_1.0.4           downloader_0.4           
FALSE  [63] rsvd_1.0.3                cellranger_1.1.0         
FALSE  [65] polyclip_1.10-0           datawizard_0.9.0         
FALSE  [67] Matrix_1.6-4              urltools_1.7.3           
FALSE  [69] base64enc_0.1-3           ggridges_0.5.3           
FALSE  [71] png_0.1-7                 parameters_0.21.3        
FALSE  [73] bitops_1.0-6              blob_1.2.1               
FALSE  [75] DelayedMatrixStats_1.10.0 qvalue_2.20.0            
FALSE  [77] jpeg_0.1-8.1              rstatix_0.7.0            
FALSE  [79] gridGraphics_0.5-1        ggsignif_0.6.0           
FALSE  [81] scales_1.3.0              memoise_1.1.0            
FALSE  [83] magrittr_2.0.1            zlibbioc_1.34.0          
FALSE  [85] compiler_4.0.2            scatterpie_0.1.5         
FALSE  [87] dqrng_0.2.1               RColorBrewer_1.1-2       
FALSE  [89] cli_3.6.1                 XVector_0.28.0           
FALSE  [91] htmlTable_2.0.1           formatR_1.8              
FALSE  [93] Formula_1.2-3             mgcv_1.8-31              
FALSE  [95] MASS_7.3-51.6             tidyselect_1.1.0         
FALSE  [97] stringi_1.4.6             forcats_0.5.1            
FALSE  [99] textshaping_0.3.6         highr_0.8                
FALSE [101] yaml_2.2.1                GOSemSim_2.14.2          
FALSE [103] BiocSingular_1.4.0        locfit_1.5-9.4           
FALSE [105] latticeExtra_0.6-29       fastmatch_1.1-0          
FALSE [107] tools_4.0.2               rio_0.5.16               
FALSE [109] rstudioapi_0.13           foreach_1.5.0            
FALSE [111] foreign_0.8-80            farver_2.0.3             
FALSE [113] ggraph_2.0.5              digest_0.6.25            
FALSE [115] rvcheck_0.1.8             BiocManager_1.30.10      
FALSE [117] shiny_1.5.0               Rcpp_1.0.7               
FALSE [119] broom_0.7.10              later_1.1.0.1            
FALSE [121] httr_1.4.2                colorspace_1.4-1         
FALSE [123] rvest_1.0.2               XML_3.99-0.5             
FALSE [125] splines_4.0.2             graphlayouts_0.7.1       
FALSE [127] ggplotify_0.0.5           systemfonts_1.0.3        
FALSE [129] xtable_1.8-4              jsonlite_1.7.2           
FALSE [131] futile.options_1.0.1      tidygraph_1.2.0          
FALSE [133] NLP_0.2-1                 R6_2.4.1                 
FALSE [135] tm_0.7-8                  Hmisc_4.4-0              
FALSE [137] pillar_1.6.4              htmltools_0.5.2          
FALSE [139] mime_0.9                  glue_1.5.1               
FALSE [141] fastmap_1.1.0             BiocParallel_1.22.0      
FALSE [143] codetools_0.2-16          fgsea_1.14.0             
FALSE [145] utf8_1.1.4                tibble_3.1.6             
FALSE [147] curl_4.3                  zip_2.0.4                
FALSE [149] GO.db_3.11.4              openxlsx_4.1.5           
FALSE [151] survival_3.2-3            rmarkdown_2.11           
FALSE [153] munsell_0.5.0             DO.db_2.9                
FALSE [155] GenomeInfoDbData_1.2.3    iterators_1.0.12         
FALSE [157] impute_1.62.0             haven_2.3.1              
FALSE [159] gtable_0.3.0              bayestestR_0.13.1