This analysis compares some gene sets raised by our study to the set of differential expressed genes in schizophrenia (SCZ) presented by (Fromer et al 2016) in their article Gene expression elucidates functional impact of polygenic risk for schizophrenia. For discussion see the post about affected genes.

## Error in library(VennDiagram): there is no package called 'VennDiagram'

Import data

diff.e.genes <- read.csv("../../data/fromer-2016-nat-neurosci/nn.4399-S5.csv", skip = 1)
our.genes <- list()
# the selected genes that were analyzed using regression
our.genes$selected <- as.character(read.csv("../../data/genes.regression.new")[[1]])
# the genes with significant association to some biological predictor
our.genes$signif <- as.character(read.csv("../../results/signif-gene-effects-either.csv")[[1]])

Results

The following genes were found to be differentially expressed by Fromer et al AND

  1. either were merely selected for analysis in our study
  2. or, furthermore, we found their parental expression bias to be significantly associated to some biological predictor
(both <- lapply(our.genes, intersect, levels(diff.e.genes$Gene.Symbol)))

## $selected
## [1] "PEG10" "IGF2" 
## 
## $signif
## [1] "PEG10"

The characteristics of the expression change for the selected genes:

diff.e.genes[diff.e.genes$Gene.Symbol %in% both$selected, c("Gene.Symbol", "logFC", "p.value")]

##    Gene.Symbol  logFC  p.value
## 44        IGF2 -0.406 1.15e-05
## 98       PEG10  0.152 5.71e-05

Further information on the significant gene(s):

read.csv("../../results/signif-gene-effects-either-manual-annot.csv", row.names = 1)[both$signif, ]

##                   Description      Gene.type Chromosome.Name
## PEG10 paternally expressed 10 protein_coding               7
##       Gene.Start..bp. rank..our.study. Associated.coefficient..our.study.
## PEG10        94656325               14                              DxSCZ
##       Phenotype     PMID
## PEG10           16341224

grid.draw(venn.diagram(list(diff.e.genes$Gene.Symbol, our.genes$selected, c("RP11-909M7.3", "PEG10", "MEST", "UBE3A")), filename=NULL, category = c("SCZ: overall expression", "called imprinted", "SCZ: parental bias"), ext.text = FALSE, cat.pos = c(-15, 15, 15), cat.cex = rep(1.2, 3), col = my.col <- c("darkgreen", "blue", "red"), fill = my.col, cat.col = my.col))

## Error in grid.draw(venn.diagram(list(diff.e.genes$Gene.Symbol, our.genes$selected, : could not find function "grid.draw"

Out of the 15 genes that our study found to be significantly associated to some biological predictor in terms of allelic bias, only 1 of them, PEG10, was/were found to be differentially expressed by Fromer et al.