Question: Trinity run failed -- Phase 1
0
gravatar for ddowlin
16 months ago by
ddowlin50
ddowlin50 wrote:

Hi all,

I am using Trinity to de novo assemble a transcriptome.

Unfortunately my run fails very early on. i.e. in Phase 1 :Clustering of RNA-Seq Reads.

The following is printed to output:

----------------------------------------------------------------------------------
-------------- Trinity Phase 1: Clustering of RNA-Seq Reads  ---------------------
----------------------------------------------------------------------------------

Converting input files. (in parallel)Thursday, July 28, 2016: 11:25:55  
CMD: gunzip -c /home/trinity/R1.fastq.gz | /home/trinityrnaseq-2.2.0/trinity-plugins/fastool/fastool --append /1 --to-fasta >> left.fa 2> /home/trinity/R1.fastq.gz.readcount 

Thursday, July 28, 2016: 11:25:55   CMD: gunzip -c /home/trinity/R2.fastq.gz | /home/trinityrnaseq-2.2.0/trinity-plugins/fastool/fastool --append /2 --to-fasta >> right.fa 2> /home/trinity/R2.fastq.gz.readcount 

Thread 2 terminated abnormally: Error, cmd: gunzip -c /home/trinity/R2.fastq.gz | /fsimb/imbc_home/trinityrnaseq-2.2.0/trinity-plugins/fastool/fastool --append /2 --to-fasta >> right.fa 2> /home/trinity/R2.fastq.gz.readcount  died with ret 256 at ./Trinity line 2206.


Thread 1 terminated abnormally: Error, counts of reads in FQ: 41119574 (as per gunzip -c /home/trinity/R1.fastq.gz | wc -l) doesn't match fastool's report of FA records: 4031700  at /home/trinityrnaseq-2.2.0/Trinity line 3087 thread 1.

main::ensure_complete_FQtoFA_conversion("gunzip -c /home/trinity/R2.fastqc.gz"..., "/home/trinity/R2.fastqc.gz) called at /home/trinityrnaseq-2.2.0/Trinity line 2116 thread 1

main::prep_seqs(ARRAY(0xc24ce0), "fq", "left", undef) called at /home/trinityrnaseq-2.2.0/Trinity line 1314 thread 1
eval {...} called at /home/trinityrnaseq-2.2.0/Trinity line 1314 thread 1


Trinity run failed. Must investigate error above.

There seems to be two errors here:

1.) Error, cmd: gunzip -c [...] died with ret 256 and

2.) Error, counts of reads in FQ: 41119574 (as per gunzip -c /home/trinity/R1.fastq.gz | wc -l) doesn't match fastool's report of FA record

Is there perhaps a problem with my fastq files?

Any help would be greatly appreciated.

Thanks in advance.

rna-seq trinity • 1.3k views
ADD COMMENTlink modified 10 weeks ago by Biostar ♦♦ 20 • written 16 months ago by ddowlin50
2

Hi,

You can ask it here: https://groups.google.com/forum/#!forum/trinityrnaseq-users

their support is great.

ADD REPLYlink written 16 months ago by Farbod3.0k
1

Before you conclude that there is a problem with your fastq files have you tried to run this a second time? Is there enough disk space available at /home/trinity to hold the uncompressed files?

ADD REPLYlink written 16 months ago by genomax39k

Thanks for the suggestion.

I unziped both fastq.gz files and re-ran Trinity.

It now gets stuck at the Jellyfish step. The specific message I am getting is:

 ----------- Jellyfish  --------------------
-- (building a k-mer catalog from reads) --
-------------------------------------------

* Running CMD: /home/trinityrnaseq-2.2.0/trinity-plugins/jellyfish/bin/jellyfish count -t 6 -m 25 -s 5936695335  --canonical  both.fa
sh: line 1: 34469 Killed                  
/home/trinityrnaseq-2.2.0/trinity-plugins/jellyfish/bin/jellyfish count -t 6 -m 25 -s 59366953$

Trinity run failed. Must investigate error above.
ADD REPLYlink modified 16 months ago • written 16 months ago by ddowlin50

Are you running this on a cluster under a job scheduler or a standalone server? How much memory do you have available?

This may indicate an available memory error since the job appears to be killed by the system now.

ADD REPLYlink modified 16 months ago • written 16 months ago by genomax39k

I am using a cluster with LSF scheduler. I requested 4 cores with 2G each.

ADD REPLYlink written 16 months ago by ddowlin50
1

Trinity requires large amount of RAM (see this and also this). 2G is not going to be enough.

If you don't have access to adequate amount of RAM locally, then consider using the galaxy server at Indiana. People have reported success using that resource for trinity here.

ADD REPLYlink modified 16 months ago • written 16 months ago by genomax39k

Thanks. I increased the memory and it seems to be running now.

ADD REPLYlink written 16 months ago by ddowlin50

You could try to run the failing commands just in your shell to get a better idea of the error message...

ADD REPLYlink written 16 months ago by WouterDeCoster24k

I ran the first command: I get the following error

terminate called after throwing an instance of 'std::runtime_error'
what():  Can't open file 'both.fa'
Aborted

I have checked the out_directory and I can open the 'both.fa' without a problem.

ADD REPLYlink written 16 months ago by ddowlin50
0
gravatar for Buffo
16 months ago by
Buffo580
Buffo580 wrote:

Try with cufflinks, it works better than Trinity: Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks Cole Trapnell, Adam Roberts, Loyal Goff, Geo Pertea, Daehwan Kim, David R Kelley, Harold Pimentel, Steven L Salzberg, John L Rinn & Lior Pachter.

ADD COMMENTlink written 16 months ago by Buffo580
1

If I'm not terribly mistaken Trinity and cufflinks don't have the same functionality... cufflinks uses aligned reads.

ADD REPLYlink written 16 months ago by WouterDeCoster24k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1431 users visited in the last hour