Why UCSC Table Browser output BED file showing multiple 3' UTR for a single isoform?

184 views
Skip to first unread message

Bharat Panwar

unread,
Jun 23, 2014, 11:03:49 AM6/23/14
to gen...@soe.ucsc.edu, Yuanfang Guan
Hello,

I am facing problem in understanding the output of UCSC Table Browser. Actually, I wanted to retrieve all the 3' UTR from the Human genome Dec. 2013 (GRCh38/hb 38) Assembly and I have got all 3'UTR information as a BED file. The problem is that it showing more than one 3' UTR for a single Isoform (e.g. Total 5 different 3'UTR for a isoform ENST00000432694.2)

chr1 172775904 172776127 ENST00000432694.2_utr3_0_0_chr1_172775905_f 0 +
chr1 173004281 173004435 ENST00000432694.2_utr3_1_0_chr1_173004282_f 0 +
chr1 173016312 173016524 ENST00000432694.2_utr3_2_0_chr1_173016313_f 0 +
chr1 173023497 173023573 ENST00000432694.2_utr3_3_0_chr1_173023498_f 0 +
chr1 173063872 173064015 ENST00000432694.2_utr3_4_0_chr1_173063873_f 0 +

I have analyzed the sequences and found that they are also very different as showing in the following web-links:

May be I am missing some information or interpreting these results wrongly but my simple query is that is it possible to have multiple 3' UTR for a single isoform?

Please help me to understand these results. 

Thanking you,

Best,
Bharat

--
Bharat Panwar
Post-Doctoral Research Fellow
GuanLab, 2044 Palmer Commons
100 Washtenaw Avenue
University of Michigan
Ann Arbor, MI - 48109
USA

Steve Heitner

unread,
Jun 24, 2014, 12:45:54 PM6/24/14
to Bharat Panwar, gen...@soe.ucsc.edu, Yuanfang Guan

Hello, Bharat.

You are correct that individual transcripts should not have multiple 3’ UTRs.  The gene you referenced is a non-coding gene and this is a product of the way the Table Browser interprets our convention for coordinates of non-coding genes.  All exons of non-coding genes on the + strand are listed as 3’ UTRs and all exons of non-coding genes on the - strand are listed as 5’ UTRs.

Thank you for bringing this to our attention.  We are now taking steps to correct this.

Please contact us again at gen...@soe.ucsc.edu if you have any further questions.  Questions sent to that address will be archived in a publicly-accessible forum for the benefit of other users.  If your question contains sensitive data, you may send it instead to genom...@soe.ucsc.edu.

---
Steve Heitner
UCSC Genome Bioinformatics Group

--

Steve Heitner

unread,
Jun 24, 2014, 1:14:16 PM6/24/14
to Bharat Panwar, gen...@soe.ucsc.edu, Yuanfang Guan

Hello again, Bharat.

Until this Table Browser issue is resolved, something that will be of help to you in listing valid 3’ UTRs is to create a filter to eliminate non-coding genes.

1. From the main Table Browser screen, on the filter line, click the “create” button

2. In the “Free-form query” box, enter the following:  cdsStart != cdsEnd

3. Click the “submit” button

This will eliminate any non-coding genes from your results.



Please contact us again at gen...@soe.ucsc.edu if you have any further questions. 

Questions sent to that address will be archived in a publicly-accessible forum for the benefit of other users.  If your question contains sensitive data, you may send it instead to genom...@soe.ucsc.edu.



---
Steve Heitner
UCSC Genome Bioinformatics Group

Reply all
Reply to author
Forward
0 new messages