Tophat Error -1: any RNA-Seq alternatives?

santanasantosl · 10-31-2012, 02:49 PM

Dear members,

I am trying to use Tophat, but it keeps failing with error -1: can not retrieve sequence XXXX from stream. Initially, I though there was some problem with the cluster I was running the analysis on. Because of the error mentioned "stream" I was suspecting of memory error, although I have more than 24GB of RAM per node. Thus, I did a test run in another cluster, and got the very same error. Well, if I run the analysis only using one thread the problem seems to resolve, however this is VERY time consuming and I have a lot of samples to analyze.
So, my suspicions is that I am seeing some bug while processing colorspace data in multiple threads.

I am wondering if any one has seem this problem and know a practical solution for that. Since it seems to be a coding problem I am not sure that there is any easy solution. So, I am wondering if someone has experience with another RNA-Seq analysis pipelines that does not involve Tophat, and could point me to some software alternatives/pipelines.

Thank you.

Lucas

adamyao · 10-31-2012, 11:47 PM

I had this problem before. I tried two different ways: to increase memory and to use single thread. My machines are 32GB and 128GB. The 32GB always gets error (out of memory) on whole human genome but the 128GB works fine. Good luck.

pbluescript · 11-01-2012, 03:46 AM

You should try STAR. http://gingeraslab.cshl.edu/STAR/
I've mentioned this elsewhere on the forums, but in the comparisons I've done, STAR yields more mapped reads, more uniquely mapped reads, and more reads with both pairs mapped than Tophat all in about 20% of the time it takes Tophat. The only downside is an increase in potential false positive splice junctions, but those can be filtered out easily enough.
It was just published too:
http://bioinformatics.oxfordjournals...ts635.abstract

santanasantosl · 11-01-2012, 08:01 AM

Adamyao I was suspecting of memory problems, so your input makes sense. Thanks
Thank you pbluescript, I will check STAR.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Bowtie error when mapping ABI RNA-seq data with Tophat	HSV-1	RNA Sequencing	5	08-01-2012 01:00 AM
Error in RNA-seq analysis of Arabidopsis following how-to- seq wiki protocol	himanshu04	RNA Sequencing	1	03-23-2012 06:58 AM
Alternatives to Nugen Ovation RNA-Seq	pbluescript	Sample Prep / Library Generation	1	02-13-2011 01:14 PM
small RNA Tool for SOLiD reads failed any alternatives?	KevinLam	Bioinformatics	4	12-27-2010 11:01 AM
error correction for RNA-seq reads	fhb	RNA Sequencing	6	11-18-2010 01:58 PM

10-31-2012, 02:49 PM	#1
santanasantosl Junior Member Location: US Join Date: Oct 2012 Posts: 2	Tophat Error -1: any RNA-Seq alternatives? Dear members, I am trying to use Tophat, but it keeps failing with error -1: can not retrieve sequence XXXX from stream. Initially, I though there was some problem with the cluster I was running the analysis on. Because of the error mentioned "stream" I was suspecting of memory error, although I have more than 24GB of RAM per node. Thus, I did a test run in another cluster, and got the very same error. Well, if I run the analysis only using one thread the problem seems to resolve, however this is VERY time consuming and I have a lot of samples to analyze. So, my suspicions is that I am seeing some bug while processing colorspace data in multiple threads. I am wondering if any one has seem this problem and know a practical solution for that. Since it seems to be a coding problem I am not sure that there is any easy solution. So, I am wondering if someone has experience with another RNA-Seq analysis pipelines that does not involve Tophat, and could point me to some software alternatives/pipelines. Thank you. Lucas

10-31-2012, 11:47 PM	#2
adamyao Member Location: Taiwan Join Date: Feb 2011 Posts: 19	I had this problem before. I tried two different ways: to increase memory and to use single thread. My machines are 32GB and 128GB. The 32GB always gets error (out of memory) on whole human genome but the 128GB works fine. Good luck.

11-01-2012, 03:46 AM	#3
pbluescript Senior Member Location: Boston Join Date: Nov 2009 Posts: 224	You should try STAR. http://gingeraslab.cshl.edu/STAR/ I've mentioned this elsewhere on the forums, but in the comparisons I've done, STAR yields more mapped reads, more uniquely mapped reads, and more reads with both pairs mapped than Tophat all in about 20% of the time it takes Tophat. The only downside is an increase in potential false positive splice junctions, but those can be filtered out easily enough. It was just published too: http://bioinformatics.oxfordjournals...ts635.abstract

11-01-2012, 08:01 AM	#4
santanasantosl Junior Member Location: US Join Date: Oct 2012 Posts: 2	Adamyao I was suspecting of memory problems, so your input makes sense. Thanks Thank you pbluescript, I will check STAR.