Create a gmt file for the Remap2020 dataset http://remap.univ-amu.fr/target_page/AATF:9606

remap2gmt.sh

#!/bin/bash
# dependancies: parallel, bedtools, pigz

# promoters
wget -N ftp://ftp.ensembl.org/pub/release-98/gtf/homo_sapiens/Homo_sapiens.GRCh38.98.gtf.gz

GTFZ=Homo_sapiens.GRCh38.98.gtf.gz

TSS=tss.bed
zcat $GTFZ | grep 'exon_number "1"' | cut -f1,4,5,7,9 \
| cut -d '"' -f1,12 | sed 's/gene_id "//' | awk '$4=="+"'  \
| awk '{OFS="\t"}{print $1,$2-1000,$2+1000,$5}' > $TSS

zcat $GTFZ | grep -w 'exon_number "1"' | cut -f1,4,5,7,9 \
| cut -d '"' -f1,12 | sed 's/gene_id "//' | awk '$4=="-"'  \
| awk '{OFS="\t"}{print $1,$3-1000,$3+1000,$5}' >> $TSS

grep -v ^MT $TSS | sed 's/^/chr/'| sortBed > $TSS.tmp && mv $TSS.tmp $TSS

# remap
wget -N http://remap.univ-amu.fr/storage/remap2020/hg38/MACS2/remap2020_all_macs2_hg38_v1_0.bed.gz

mkdir peaks && cd peaks

ulimit -n  1000000

zcat ../remap2020_all_macs2_hg38_v1_0.bed.gz  | awk '{print > $4".bed"}'

pk2gene(){
PEAK=$1
GENES=$2
  bedtools intersect -wo -a $GENES -b $PEAK \
  | sort -k9gr | awk '!arr[$4]++ {print $4}' \
  | paste -s -d '\t' \
  | sed "s/^/${PEAK}\tFromRemap2020\t/" > $PEAK.gmt
}
export -f pk2gene
parallel pk2gene ::: *bed ::: ../tss.bed

cat *gmt | pigz >../remap2020.gmt.gz

The gmt file is available from
http://dee2.io/data/tmp/

Could have saved a bunch of time if I used GTFtools to extract TSS coordinates!!

gtftools.py -t Homo_sapiens.GRCh38.94.gtf.tss.bed -w 1000  Homo_sapiens.GRCh38.94.gtf