SEQanswers

Go Back   SEQanswers > Applications Forums > RNA Sequencing



Similar Threads
Thread Thread Starter Forum Replies Last Post
[DEXSeq] exon counts to "PSI" (exon inclusion level) yerbol Bioinformatics 3 11-23-2015 05:32 PM
edgeR spliceVariants: gene- and exon-level dispersion bstephenwhite@gmail.com Bioinformatics 3 09-11-2014 03:47 PM
DEXSeq - Counting with HT-seq at the exon level Yohann Bioinformatics 1 05-09-2014 11:35 AM
Gene expression level from replicates gen2prot Bioinformatics 4 07-10-2013 02:03 AM
How to compute exression ratio of Human Exon 1.0 ST array on probe level tujchl Bioinformatics 14 10-30-2012 07:30 AM

Reply
 
Thread Tools
Old 06-10-2014, 05:14 AM   #1
Ohad
Member
 
Location: Israel TA

Join Date: Jul 2013
Posts: 28
Default exon expression level

Hi there.

I would like to to get for a list of exons their expression level from my RNA-seq data.
I thought about using cufflinks but the problem is that cufflinks generate FPKM at the isoform level. I usually feed cufflinks with a refseq gtf file downloaded from UCSC table browser (under genes and genes prediction), and cufflinks use that to build locus.
I thought to switch back from locus to exons (using the ref ID NM/R_somenumber) but it's a wrong way to go by since:

1) many exons participate in more than one transcript, and each has a different FPKM value
2) FPKM number are probably calculated considering fragment length and each transcripts has its own, so I cannot just sum up the FPKM from all transcripts for a particular exon

If I was in a world in which every gene had only one transcripts I guess my life would be easier and I would be blonde and handsome , but reality is crueler than that.

All tables I can download on ucsc include variants so I got my hands on illumina truseq exome file containing all exons per gene in a definitively matter, meaning each exon appear once , even if some belongs to different variants. But it seems this file is no good for cufflinks and it was not able to build locus right.

I thought of just using samtools for each exon:
samtools view accpeted_hits.bam chr1:exonstrat-exonend | wc -l
and just get and number and calculate my own FPKM using the exon's length (and total number of reads)

Do you think using samtools is fine ?
Ohad is offline   Reply With Quote
Old 06-10-2014, 05:24 AM   #2
dpryan
Devon Ryan
 
Location: Freiburg, Germany

Join Date: Jul 2011
Posts: 3,480
Default

Just use DEXseq and its python script for counting exonic bins. This is a solved problem. Yes, you'll get raw counts, but you can do whatever you want with those then.
dpryan is offline   Reply With Quote
Old 06-10-2014, 05:28 AM   #3
Ohad
Member
 
Location: Israel TA

Join Date: Jul 2013
Posts: 28
Default

Do I need to install the whole package or can I just use this one script ?
Ohad is offline   Reply With Quote
Old 06-10-2014, 05:36 AM   #4
dpryan
Devon Ryan
 
Location: Freiburg, Germany

Join Date: Jul 2011
Posts: 3,480
Default

As I recall, the python scripts are just wrappers around htseq-count, so you can likely use them without the rest of the R package. Having said that, if you're interested in using the data to look at differential exon usage then DEXseq would be the tool to use anyway.
dpryan is offline   Reply With Quote
Old 06-10-2014, 05:45 AM   #5
Ohad
Member
 
Location: Israel TA

Join Date: Jul 2013
Posts: 28
Default

As for now I only need the expression, will samtools be enough ?
I would like to save the time of learning how to use this package as I'm not that familiar yet with R
Ohad is offline   Reply With Quote
Old 06-10-2014, 05:53 AM   #6
dpryan
Devon Ryan
 
Location: Freiburg, Germany

Join Date: Jul 2011
Posts: 3,480
Default

As long as you don't have paired-end reads, then yes, that'll work. You could also just
Code:
samtools view -c alignments.bam chr1:start-end
since the -c option will do the counting for you.
dpryan is offline   Reply With Quote
Old 06-10-2014, 05:59 AM   #7
Ohad
Member
 
Location: Israel TA

Join Date: Jul 2013
Posts: 28
Default

ok thank you for this help.

I do have paired-end reads, but I guess I could write some script to handle that, using the reads ID.

Nevertheless, I do preform differential exons analysis quite often, as I study alternative splicing, so it's about time I learn to use DEXseq anyhow.
Ohad is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 01:16 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO