Re: [trinityrnaseq-users] trinity assembly taking a long time in chrysalis

384 views

Skip to first unread message

Brian Haas

unread,

Aug 4, 2017, 8:35:40 AM8/4/17

to Katherine Dougan, trinityrnaseq-users

Hi Katherine,

With large data sets, you'd need to use the --normalize_reads parameter. This is now on by default in the latest version of Trinity v2.4.0, but you may need to invoke it with v2.3.2

If you decide to rerun it, you'll need to start in a fresh workspace and either delete all existing outputs or run in a clean new working directory, so Trinity won't try to reuse any preexisting outputs from the current run.

best,

~brian

On Thu, Aug 3, 2017 at 9:21 PM, Katherine Dougan <katherine...@gmail.com> wrote:

Below is the run log for a de novo transcriptome assembly with ~600 million PE reads. It has been stuck on this last command for about 48 hours or so now. I have assembled several transcriptomes and haven't seen any of them take this long in this step. Anyone else run into anything familiar? I just want to make sure that it has not stalled and I'm wasting my hours on my university's HPC. Thanks!

____________________________________________________________________

Trinity version: Trinity-v2.3.2
-ERROR: couldn't run the network check to confirm latest Trinity software version.

Wednesday, August 2, 2017: 08:56:35 CMD: java -Xmx64m -XX:ParallelGCThreads=2 -jar /home/applications/trinity/2.3.2/util/support_scripts/ExitTester.jar 0
Wednesday, August 2, 2017: 08:56:35 CMD: java -Xmx64m -XX:ParallelGCThreads=2 -jar /home/applications/trinity/2.3.2/util/support_scripts/ExitTester.jar 1
Wednesday, August 2, 2017: 08:56:35 CMD: mkdir -p /scratch/kdoug023/PRJNAtrenchii/trinity_out
Wednesday, August 2, 2017: 08:56:35 CMD: mkdir -p /scratch/kdoug023/PRJNAtrenchii/trinity_out/chrysalis

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

Converting input files. (in parallel)Wednesday, August 2, 2017: 08:56:35 CMD: ln -s /scratch/kdoug023/trenchii_transcriptome/trenchii_final_forward_reads.fasta left.fa
Wednesday, August 2, 2017: 08:56:35 CMD: /home/applications/trinity/2.3.2/util/support_scripts/revcomp_fasta.pl /scratch/kdoug023/trenchii_transcriptome/trenchii_final_reverse_reads.fasta >> right.fa
Wednesday, August 2, 2017: 09:18:00 CMD: touch left.fa.ok right.fa.ok
Wednesday, August 2, 2017: 09:18:00 CMD: cat left.fa right.fa > /scratch/kdoug023/PRJNAtrenchii/trinity_out/both.fa
Wednesday, August 2, 2017: 09:20:20 CMD: touch /scratch/kdoug023/PRJNAtrenchii/trinity_out/both.fa.ok
-------------------------------------------
----------- Jellyfish --------------------
-- (building a k-mer catalog from reads) --
-------------------------------------------

* Running CMD: /home/applications/trinity/2.3.2/trinity-plugins/jellyfish/bin/jellyfish count -t 15 -m 25 -s 197243297753 /scratch/kdoug023/PRJNAtrenchii/trinity_out/both.fa
* Running CMD: /home/applications/trinity/2.3.2/trinity-plugins/jellyfish/bin/jellyfish dump -L 1 mer_counts.jf > jellyfish.kmers.fa
* Running CMD: /home/applications/trinity/2.3.2/trinity-plugins/jellyfish/bin/jellyfish histo -t 15 -o jellyfish.kmers.fa.histo mer_counts.jf
----------------------------------------------
--------------- Inchworm ---------------------
-- (Linear contig construction from k-mers) --
----------------------------------------------

* Running CMD: /home/applications/trinity/2.3.2/Inchworm/bin//inchworm --kmers jellyfish.kmers.fa --run_inchworm -K 25 -L 25 --monitor 1 --num_threads 15 --PARALLEL_IWORM > /scratch/kdoug023/PRJNAtrenchii/trinity_out/inchworm.K25.L25.fa.tmp
* Running CMD: mv /scratch/kdoug023/PRJNAtrenchii/trinity_out/inchworm.K25.L25.fa.tmp /scratch/kdoug023/PRJNAtrenchii/trinity_out/inchworm.K25.L25.fa
Wednesday, August 2, 2017: 14:41:27 CMD: touch /scratch/kdoug023/PRJNAtrenchii/trinity_out/inchworm.K25.L25.fa.finished
--------------------------------------------------------
-------------------- Chrysalis -------------------------
-- (Contig Clustering & de Bruijn Graph Construction) --
--------------------------------------------------------

inchworm_target: /scratch/kdoug023/PRJNAtrenchii/trinity_out/both.fa
bowite_reads_fa: /scratch/kdoug023/PRJNAtrenchii/trinity_out/both.fa
chrysalis_reads_fa: /scratch/kdoug023/PRJNAtrenchii/trinity_out/both.fa
* Running CMD: /home/applications/trinity/2.3.2/util/misc/fasta_filter_by_min_length.pl /scratch/kdoug023/PRJNAtrenchii/trinity_out/inchworm.K25.L25.fa 100 > /scratch/kdoug023/PRJNAtrenchii/trinity_out/chrysalis/inchworm.K25.L25.fa.min100
* Running CMD: bowtie2-build -o 3 /scratch/kdoug023/PRJNAtrenchii/trinity_out/chrysalis/inchworm.K25.L25.fa.min100 /scratch/kdoug023/PRJNAtrenchii/trinity_out/chrysalis/inchworm.K25.L25.fa.min100 1>/dev/null
* Running CMD: bash -c " set -o pipefail;bowtie2 --local -a --threads 15 -f -x /scratch/kdoug023/PRJNAtrenchii/trinity_out/chrysalis/inchworm.K25.L25.fa.min100 /scratch/kdoug023/PRJNAtrenchii/trinity_out/both.fa | samtools view -@ 15 -F4 -Sb - | samtools sort -m 53687091200 -@ 15 -no - - > /scratch/kdoug023/PRJNAtrenchii/trinity_out/chrysalis/iworm.bowtie.nameSorted.bam"

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.
To post to this group, send email to trinityrnaseq-users@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.