Trinity Rna-Seq De Novo: Chrysalis Died With Ret 11

Question

[Pasta]

Hi there,

I would like to assemble RNA-seq data de novo using Trinity. I got paired-end reads (44 million). The problem is that I keep having this error message:

Error, cmd: /usr/local/bin/trinityrnaseq_r2011-07-13/Chrysalis/Chrysalis -i both.fa -iworm   inchworm.K25.L48.DS.fa -o chrysalis  -min 200 -dist 300  -butterfly /usr/local/bin/trinityrnaseq_r2011-07-13/Butterfly/Butterfly.jar  died with ret 11 at Trinity.pl line 459.

Error 11 being "Try Again" in Linux.

This is the command I ran:

sudo perl Trinity.pl --seqType fa --left No_Hits_reads1.fasta --right No_Hits_reads2.fasta --CPU 6 --run_butterfly --bflyHeapSpace 10G

As suggested in the documentation I set the stacksize to unlimited (ulimit -s unlimited), as you can see below:

core file size          (blocks, -c) 0
data seg size           (kbytes, -d) unlimited
scheduling priority             (-e) 20
file size               (blocks, -f) unlimited
pending signals                 (-i) 16382
max locked memory       (kbytes, -l) 64
max memory size         (kbytes, -m) unlimited
open files                      (-n) 1024
pipe size            (512 bytes, -p) 8
POSIX message queues     (bytes, -q) 819200
real-time priority              (-r) 0
stack size              (kbytes, -s) unlimited
cpu time               (seconds, -t) unlimited
max user processes              (-u) unlimited
virtual memory          (kbytes, -v) unlimited
file locks                      (-x) unlimited

Do you guys have any idea on how to solve this problem ? I have "only" 24 Gb of RAM, could that be a problem ?

Thanks+++

Answer 1

[Avian]

Although I cannot rule out a specific error with Trinity, I think it is highly likely that your issues do indeed relate to the amount of RAM that you have available.

Although it is only a rough guide, it is suggested that Trinity requires approximately 1gb of RAM for every million reads. Your 44 million reads are likely to cause problems if you only have 24gb RAM available.

If possible, you might want to find a machine with more RAM to run the Inchworm and Chrysalis parts of Trinity. RAM is less of an issue for Butterfly and should work fine on your 24gb machine.

You might also want to use fewer processors, as this might lower the memory footprint. By selecting 6 cores and a butterfly heap space of 10Gb, you are effectively stating that you have 60gb of RAM available.

Finally, trinity was updated recently with a number of efficiency improvements, so you may have more success with the updated version.

[Pasta]

Thanks for your answer. I think my problem is definitely linked to RAM as you suggested. I will try your trick.

Answer 2

[Quinnie]

I've faced the same problems too, with 40million pair-end reads, by setting the Stack size to unlimited, the error still arise although I work on 60Gb RAM server.

[Pasta]

I wonder what is the recommended amount of RAM for 40M PE reads ?

Answer 3

[Marina Manrique]

Hi Pasta,

I haven't used Trinity so far so I can't help you with this error, but maybe you could try to use Oases http://www.ebi.ac.uk/~zerbino/oases/, a version of Velvet for transcriptome data