Gene list from bed file

Question

[o.hickman]

I have a .bed file with Chromosomal coordinates instead of gene names in columns 1, 2 and 3. How do I get the coesponding gene list from this information?

chr7 120792856 120792910 K562_rep01 1000 + 5.486570215 14.23664891 -1 -1

Thanks in advance!!

Oliver

[GenoMax]

Is the gene name in your file itself? Though KHDRBS1 may be on chr1 than chr7 if that line is correct.

[o.hickman]

Hi, Sorry, that column is the name of the experiment. The gene name is definitely not in there. Can you help? O.

[zjhzwang]

You can use biomaRt package to get the genes' chromesome information, then you can map your bed file to gene symbols or gene IDs.

Answer 1

[e.rempel]

Hi, you also can use bedtools intersect. From the online manual:

bedtools intersect allows one to screen for overlaps between two sets of genomic features.

One set is your bed file, the second file could be a GFF/GTF file with the gene model of your organism.

Answer 2

[agata88]

I think you can do it by UCSC Table Browser:

https://genome.ucsc.edu/

Select group: Genes and Genes predictions, RefSeq, hg19/hg38 Define regions in region field (chr:start-end), then select output format: all fields from selected table, get output.

At the end of text file you will have field name2 - gene name.

I think it is possible to filter all unnecessary informations from this file and leave only gene name, transcript etc. and save in BED file.

Best,

Agata

[Thind amarinder]

I tried this and I would like to add something extra. Also select Track:- UCSC. I obtained multiple lines w.r.t. original co-ordinates in .bed so I collapsed it using bedtools: bedtools map -a original.bed -b ucsc_based_gene_list.bed -c 5 -o collapse

Answer 3

[badribio]

Post processing and filtering redundancies, You could try bedmap as explained by @ Alex Reynolds in this post C: how to get gene names and gene id given co ordinates from ensembl its fast and elegant.