Bedfile format deeptools

Question

[Lalita]

Hello everyone,

Using deeptools I am trying to plot the enrichment in Genebody and TSS for this we have to give one BED file region of interest in computeMatrix. I am bit confused about the format based on a thread in biostar TSS metaprofile using Deeptoolsbased on this thread I have converted my bed file trascript wise to gene wise can anyone suggest me is it a way of doing?

Original bed file transcript wise bed file

1   4970856 4971128 ENSMUSG00000085623  +
1   4971631 4971828 ENSMUSG00000085623  +
1   4973564 4976820 ENSMUSG00000085623  +
1   6213292 6213347 ENSMUSG00000090031  -
1   6214232 6214590 ENSMUSG00000090031  -
1   4879201 4880669 ENSMUSG00000062588  -
1   4880048 4880051 ENSMUSG00000062588  -
1   4880051 4880651 ENSMUSG00000062588  -
1   4880648 4880651 ENSMUSG00000062588  -

convert Gene wise bed file

1   4970856 4976820 ENSMUSG00000085623  +
1   6213292 6214590 ENSMUSG00000090031  -
1   4879201 4880669 ENSMUSG00000062588  -

Answer 1

[Devon Ryan]

A TSS is a transcript-level feature, so normally you would have a BED file with an entry for each transcript. Ideally the transcript ID would appear in column 4.

[Lalita]

Devon Ryan Thank you for the response, than no need to do the gene wise bed file. I will replace gene ID with Transcript ID in the original bed file.

[Lalita]

Devon Ryan I am little bit confused because in deeptools documentation its given that gene wise start and end deeptools doc please clarify

[Devon Ryan]

You can use a genewise start, the question is whether that makes sense for your use-case. deepTools doesn't care what the regions represent, they could be any random regions of interest. But if you're interested in looking at transcription start sites then you'll normally want to use transcript coordinates and IDs.