Region-set enrichment analysis using LOLA

library("LOLA")
library("GenomicRanges")
library("Hmisc")
library("pheatmap")

An example using MHBs from adrenal gland

MHB="BED/Adrenal_MHB.bed"

# MHB regions
background="BED/MHB_Cat.bed"
mhb=readBed(MHB)

# Universe_Sets Background
universe_set=readBed(background)

# Overlaped Regions 
states=loadRegionDB("BED/hg19_test")
    
# Runing LOLA
locResults = runLOLA(mhb, universe_set, states, cores=5)

constant = 17
locResults$qValue = (10^(-locResults$pValueLog)) * (constant / locResults$rnkPV)
head(locResults);dim(locResults)

A data.table: 6 × 23

userSet	dbSet	collection	pValueLog	oddsRatio	support	rnkPV	rnkOR	rnkSup	maxRnk	⋯	d	description	cellType	tissue	antibody	treatment	dataSource	filename	size	qValue
<int>	<int>	<chr>	<dbl>	<dbl>	<int>	<int>	<int>	<int>	<int>	⋯	<int>	<chr>	<chr>	<chr>	<chr>	<chr>	<chr>	<chr>	<dbl>	<dbl>
1	2	ucsc_example	210.191253	2.185228	3529	1	1	1	1	⋯	69405	ucsc_example	NA	NA	NA	NA	NA	Adrenal_H3K27ac.bed	158566	1.094450e-209
1	10	ucsc_example	58.564747	1.537888	2500	2	2	11	11	⋯	75085	ucsc_example	NA	NA	NA	NA	NA	Ovary_H3K27ac.bed	194882	2.315642e-58
1	1	ucsc_example	10.963068	1.188053	2707	3	3	10	10	⋯	66551	ucsc_example	NA	NA	NA	NA	NA	Adipose_H3K27ac.bed	123493	6.169644e-11
1	13	ucsc_example	10.092572	1.174367	3167	4	4	4	4	⋯	59554	ucsc_example	NA	NA	NA	NA	NA	Spleen_H3K27ac.bed	130465	3.434132e-10
1	14	ucsc_example	8.211183	1.153098	3454	5	5	2	5	⋯	54817	ucsc_example	NA	NA	NA	NA	NA	Stomach_H3K27ac.bed	135084	2.090721e-08
1	17	ucsc_example	6.673084	1.135736	3124	6	6	5	6	⋯	59348	ucsc_example	NA	NA	NA	NA	NA	Thyroid_H3K27ac.bed	142299	6.014693e-07

1. 17
2. 23

# mask non-significant results
log=locResults$qValue>0.05
locResults$rnkPV[log]<-NA
result=locResults[,c('rnkPV','filename')];result

A data.table: 17 × 2

rnkPV	filename
<int>	<chr>
1	Adrenal_H3K27ac.bed
2	Ovary_H3K27ac.bed
3	Adipose_H3K27ac.bed
4	Spleen_H3K27ac.bed
5	Stomach_H3K27ac.bed
6	Thyroid_H3K27ac.bed
7	Lung_H3K27ac.bed
8	Heart_H3K27ac.bed
9	Liver_H3K27ac.bed
NA	Esophagus_H3K27ac.bed
NA	Pancreas_H3K27ac.bed
NA	Colon_H3K27ac.bed
NA	Thymus_H3K27ac.bed
NA	Breast_H3K27ac.bed
NA	Placenta_H3K27ac.bed
NA	Bcell_H3K27ac.bed
NA	Tcell_H3K27ac.bed

Enrichment of MHBs in H3K27ac ChIP-seq peaks

Result=read.table("BED/Result_H3K27ac.txt",sep="\t",header = T)
colnames(Result)=(sapply(strsplit(colnames(Result), "\\."), function(z) z[1]))
colnames(Result)=capitalize(colnames(Result))
rownames(Result)=Result$Filename
Result=Result[,-1]
rownames(Result)=capitalize(rownames(Result))
colnames(Result)=(sapply(strsplit(colnames(Result), "\\_"), function(z) z[1]))
head(Result)

A data.frame: 6 × 17

	Adipose	Adrenal	B	Breast	Colon	Esophagus	Heart	Liver	Lung	Ovary	Pancreas	Placenta	Spleen	Stomach	T	Thymus	Thyroid
	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>	<int>
Adipose	2	3	9	4	17	9	1	17	6	5	17	17	8	17	7	17	17
Adrenal gland	5	1	14	17	17	17	2	17	13	3	17	17	6	17	15	17	17
B-cell	13	17	2	17	2	17	17	17	8	17	17	17	2	17	2	2	17
Breast	17	17	4	1	9	6	17	17	11	17	17	17	17	17	5	17	17
Colon	6	17	11	17	1	10	17	17	3	17	17	17	7	3	6	17	17
Esophagus	9	17	10	2	5	1	17	17	7	17	17	17	17	5	11	17	17

Heatmap

options(repr.plot.width = 8, repr.plot.height=6)
pheatmap(Result,
    show_rownames=T,main="H3K27ac Enrichment rnkPV",
    cluster_cols= F,angle_col=45,fontsize_row=11,cellwidth=21,
    cluster_rows=F,cellheight=18,fontsize_col=12,
    how_colnames = T,
    legend_breaks = c(1,17),legend_labels = c("1","17"),
    scale = "none",border_color = "grey90", display_numbers = matrix(ifelse(Result > 1, "", "*"), 
                                                                     nrow = nrow(Result)),
    color = colorRampPalette(colors = c("#F71E35", "#FCD271","#fcebb6","white"))(50))