I want to map my SNP positions to gff file and extract the mapped gene name, gene coordinates.
I have two files, In one file, I have chromosome number and SNP position and another file which is gff file. I want to find if the SNPs are located inside genes and extract gene name and gene coordinates from gff file. Is there some CLI-tool where you can pass a SNP position with chromosome number and a gff file and it returns the name of the genes with coordinates where SNP mapped?
1. file1 (SNP informtion)
Chrm_num SNP position
1 45106203
1 143047631
1 146650701
1 149261376
2 47602826
2 134579820
3 21063116
3 24559129
3 28000146
3 34244849
3 56395014
3 66228111
3 88703762
3 89967513
2. file 2 (gff file)
1 RefSeq region 1 161428367 . + . ID=id0;Dbxref=taxon:9913;Name=1;breed=Hereford;chromosome=1;gbkey=Src;genome=chromosome;mol_type=genomic DNA
1 Gnomon gene 26068 28997 . + . ID=gene0;Dbxref=GeneID:104976623;Name=LOC104976623;gbkey=Gene;gene=LOC104976623;gene_biotype=pseudogene;pseudo=true
1 Gnomon exon 26068 28997 . + . ID=id1;Parent=gene0;Dbxref=GeneID:104976623;gbkey=exon;gene=LOC104976623
Please let me know, I tried it very hard but I am not getting it.
•
written
by
shachigahoimbi •
one way is to use the intersect function from BedTools.
awk 'BEGIN{FS=OFS="\t"}{if(NR>1){ print $1,$2-1,$2 }}' snps.txt |
bedtools intersect -wb -a genes.gtf -b stdin | awk '$3=="gene"'
From the output you can decide what columns to keep.
Also, if you want to get the entire gene coordinates instead of just the overlapping bases then change '-wb' to -'wo'
•
written
by
husensofteng
Thank you so much for telling it in detail. I got the result exactly what I want. Thanks once again!
If an answer was helpful you should upvote it, if the answer resolved your question you should mark it as accepted.
Log in to answer this question.
Welcome to biostars. I have two suggestions :-)
We like to put you on the right track, but then it is easier if you show us what you already tried. Maybe you are almost there and we just have to point you in the right direction for the final step.
Second, it would probably be easier if you have your variants in vcf format, since for those files many tools are developed. Do you?