[DEXSeq] exon counts to "PSI" (exon inclusion level)

yerbol · 02-12-2013, 02:42 AM

Hello all,

I`m using DEXSeq to find differential splicing. It operates with raw counts of reads aligned to the exonic segments. But frequently it is worth to have inclusion levels ("PSI") assigned to exons, which is proportion of transcripts including given exonic segment among all transcripts of given gene.

Calculation of absolute expression level of exon is seems to be not very difficult (something like FPKM): exon count * size factor (or divided to the size of library) / length of exon.

But for inclusion level we need to find ratio of expression level of exon (FPKM/counts normalized to length) to the expression level of whole gene:
something like "PSI" = "exonFpkm"/"geneFpkm".

Is there any smart way to calculate something like PSI from exon count objects used in DEXSeq?

Thank you.

DerSeb · 10-08-2015, 11:18 PM

Hi yerbol,

we have done just that:

http://onlinelibrary.wiley.com/doi/1...16s87/abstract

There is also a script available that will calculate the PSI scores as explained in the protocol for you. You can either use count files or read files as input.

All the best,
Sebastian

Neuromancer · 11-23-2015, 07:00 AM

Hi Sebastian, any chance you could provide fulltext to the article you linked? Can't get it though my university....

Thanks,

Roman

DerSeb · 11-23-2015, 06:32 PM

Dear Roman,

feel free to contact me at:

sebastian.schaefer and at the domain nhcs.com.sg

Thanks for your interest in our work,
Sebastian

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
DEXSeq error in estimateDispersions: match.arg(start.method, c("log(y)", "mean"))	fpadilla	Bioinformatics	14	07-03-2013 03:11 PM
DEXSeq gene level counts	Julien Roux	Bioinformatics	3	11-28-2012 01:31 AM
Using Cufflinks to calculate cassette exon inclusion rate	chaodai	Bioinformatics	1	07-30-2012 05:40 AM
Getting Gene-Level "Expression" Value	srp33	RNA Sequencing	1	10-08-2010 02:58 PM
"Systems biology and administration" & "Genome generation: no engineering allowed"	seb567	Bioinformatics	0	05-25-2010 01:19 PM

02-12-2013, 02:42 AM	#1
yerbol Junior Member Location: Russia Join Date: Feb 2012 Posts: 7	[DEXSeq] exon counts to "PSI" (exon inclusion level) Hello all, I`m using DEXSeq to find differential splicing. It operates with raw counts of reads aligned to the exonic segments. But frequently it is worth to have inclusion levels ("PSI") assigned to exons, which is proportion of transcripts including given exonic segment among all transcripts of given gene. Calculation of absolute expression level of exon is seems to be not very difficult (something like FPKM): exon count * size factor (or divided to the size of library) / length of exon. But for inclusion level we need to find ratio of expression level of exon (FPKM/counts normalized to length) to the expression level of whole gene: something like "PSI" = "exonFpkm"/"geneFpkm". Is there any smart way to calculate something like PSI from exon count objects used in DEXSeq? Thank you.

10-08-2015, 11:18 PM	#2
DerSeb Member Location: Singapore Join Date: Oct 2009 Posts: 44	Hi yerbol, we have done just that: http://onlinelibrary.wiley.com/doi/1...16s87/abstract There is also a script available that will calculate the PSI scores as explained in the protocol for you. You can either use count files or read files as input. All the best, Sebastian

11-23-2015, 07:00 AM	#3
Neuromancer Member Location: Goettingen, Germany Join Date: Aug 2011 Posts: 28	Hi Sebastian, any chance you could provide fulltext to the article you linked? Can't get it though my university.... Thanks, Roman

11-23-2015, 06:32 PM	#4
DerSeb Member Location: Singapore Join Date: Oct 2009 Posts: 44	Dear Roman, feel free to contact me at: sebastian.schaefer and at the domain nhcs.com.sg Thanks for your interest in our work, Sebastian