extract gene TSS/TES site form gencode/ensembl/gencode database GTF file and export bed format file.
$ pip install GetTsite
help infomation:
$ GetTss -h
usage: GetTss --database ucsc --gtffile hg19.ncbiRefSeq.gtf --tssfile testTSS.bed
Get gene TSS site and export bed format from GTF annotation file.
optional arguments:
-h, --help show this help message and exit
-v, --version show program's version number and exit
-d {ucsc,ensembl,gencode}, --database {ucsc,ensembl,gencode}
which annotation database you choose. (default="ensembl")
-g GTFFILE, --gtffile GTFFILE
input your GTF file. (ucsc/ensembl/gencode)
-t TSSFILE, --tssfile TSSFILE
output your TSS file. (test-TSS.bed)
Thank your for your support, if you have any questions or suggestions please contact me: 3219030654@stu.cpu.edu.cn.
for ucsc gtf file:
$ GetTss -d ucsc -g hg19.ncbiRefSeq.gtf -t ucsc-TSS.bed
Your job is starting, please wait!
You GTF file have: 104178 transcripts.
Your task has down!
$ head -n 3 ucsc-TSS.bed
chrMT 16023 16024 TRNP . -
chrMT 15887 15888 TRNT . +
chrMT 14746 14747 CYTB . +
for gencode/ensembl gtf file:
$ GetTss -d gencode -g gencode.v19.annotation.gtf -t test-TSS.bed
Your job is starting, please wait!
You GTF file have: 57820 genes.
Your task has down!
$ head -n 3 test-TSS.bed
chr1 11868 11869 ENSG00000223972.4 . +
chr1 29806 29807 ENSG00000227232.4 . -
chr1 29553 29554 ENSG00000243485.2 . +
the usage of GetTes is same as GetTss:
$ GetTes -d ucsc -g hg19.ncbiRefSeq.gtf -t ucsc-TSS.bed
$ GetTes -d gencode -g gencode.v19.annotation.gtf -t test-TES.bed
compute matrix:
$ computeMatrix reference-point -S normal.bw treat.bw \
-R myTSS.bed \
--referencePoint center \
-a 3000 -b 3000 -p 25 \
-out matrix.tab.gz
plot Profile:
$ plotProfile -m matrix.tab.gz \
-out profile.pdf \
--perGroup \
--plotTitle 'test profile'