Trinity unusual error

Question: Trinity unusual error

19 months ago by

Hi all!

I got twice the same error by running Trinity on Galaxy, that is:

Thread 1 terminated abnormally: Error, counts of reads in FQ: 40001112.25 (as per cat /pylon5/mc48nsp/xcgalaxy/main_staging//15659867/inputs/dataset_19697450.dat | wc -l) doesn't match fastool's report of FA records: 40001113 at /opt/packages/trinity/2.2.0/Trinity line 3087 thread 1.

As a consequence Trinity stops producing an empty file.

As far as I know that seems to be something related with the number of reads inside the loaded files but I have no idea how to deal with that (it looks more like a Trinity bug on Galaxy). That is, in my opinion, quite unusual because the same workflow has been running multiple times with many samples from the same experiment.

N.B. --> The same error is given independently from two samples processed separately.

Anyone can give me some tips?

Thank you,

Luca

rna-seq software error trinity assembly galaxy • 906 views

ADD COMMENT • link •

modified 15 months ago by josefinagutierrezd • 10 • written 19 months ago by luca.marisaldi • 30

Hello, We are testing Trinity again. Would you also please send in a bug report so that we can look at the error within the context of the experiment? Please leave the inputs and error datasets undeleted and include a link to this post. This does look like a server side issue: https://galaxyproject.org/support/tool-error/#determining-the-job-failure-type

More feedback soon.

Thanks, Jen, Galaxy team

ADD REPLY • link written 19 months ago by Jennifer Hillman Jackson ♦ 25k

Hi Jen, well I run the job twice more after posting the question without changing anything (simply using the icon "run this job again")

The last attempt was successful and Trinity proceeded further completing the run properly in both samples.

For this reason I lost the info regarding the error. If it is going to happen again I proceed by submitting the full report as you suggested.

Thank you for helping us!

Bye,

Luca

ADD REPLY • link written 19 months ago by luca.marisaldi • 30

19 months ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hi Luca,

The connection with Bridges (where Trinity is run) has been problematic on and off for about a week. So, no need to send in the error if it is resolved now for you.

Should this come up again, we want to know, from you and anyone else who runs into a problem.

Thanks for the followup! Jen

ADD COMMENT • link written 19 months ago by Jennifer Hillman Jackson ♦ 25k

15 months ago by

josefinagutierrezd • 10

josefinagutierrezd • 10 wrote:

Hi,

I'm having the same error as Luca using Trinity. Just wanted to check if there is something to do or is the server that is failing again?

My output is an empty file and this log:

Thread 1 terminated abnormally: Error, counts of reads in FQ: 708580 (as per cat /pylon5/mc48nsp/xcgalaxy/main_staging//16559348/inputs/dataset_20921519.dat | wc -l) doesn't match fastool's report of FA records: 1417160 at /opt/packages/trinity/2.2.0/Trinity line 3087 thread 1.

Thread 2 terminated abnormally: Error, counts of reads in FQ: 708580 (as per cat /pylon5/mc48nsp/xcgalaxy/main_staging//16559397/inputs/dataset_20921521.dat | wc -l) doesn't match fastool's report of FA records: 1417160 at /opt/packages/trinity/2.2.0/Trinity line 3087 thread 2.

Thanks!

Josefina

ADD COMMENT • link written 15 months ago by josefinagutierrezd • 10

Hi Josefina, There is likely a problem with the inputs. I have a copy of your data and am running some tests. I have a pretty good idea about how to make a correction (two actually) but will send full feedback once the test jobs complete. Jen, Galaxy team

ADD REPLY • link written 15 months ago by Jennifer Hillman Jackson ♦ 25k

The working solution is to use the original fastq datasets as input instead of the fasta converted input.

After review and tests, I do not believe that the Trinity service set up at http://usegalaxy.org is able to handle fasta input. I'll be creating an enhancement ticket to address this once all tests complete: either filter inputs so that ONLY fastq is recognized as accepted input or adjust the wrapper so that when fasta is given as input, that is communicated to the server that processes the jobs. In addition, failed jobs like this should be "red" in the history, not green with empty output with an error message describing the failure in the logs.

The service is somewhat new and still in Beta - so thanks for reporting this! Jen, Galaxy team

ADD REPLY • link written 15 months ago by Jennifer Hillman Jackson ♦ 25k

This is the ticket to address the usage issue. It can be followed, comments can be added by anyone with a Github account, and thumbs-up votes show community interest. https://github.com/galaxyproject/galaxy/issues/4418

ADD REPLY • link written 15 months ago by Jennifer Hillman Jackson ♦ 25k

19 months ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Update:

Trinity was corrected again mid-day Friday 4/28/17 EST. Please re-run any jobs that failed.

If you have jobs that were started after that time, those will also need to be re-run (and the original grey jobs deleted/purged).

Sorry for the inconvenience, Jen, Galaxy team

ADD COMMENT • link written 19 months ago by Jennifer Hillman Jackson ♦ 25k

18 months ago by

ilange • 0

ilange • 0 wrote:

Hi Jennifer, I have the same error message as Luca Although I was able to run an almost identical trinity assembly in parallel on our cluster and it finished. I rerun the failed assembly once. No luck. Here is the error:

Thread 2 terminated abnormally: Error, counts of reads in FQ: 5045646.25 (as per cat //RNA_assemblies/a/trimmed_SRR515_2.fastq | wc -l) doesn't match fastool's report of FA records: 5045647 at /opt/apps/trinity/2.2.0/Trinity line 3087 thread 2. main::ensure_complete_FQtoFA_conversion('cat /data/cahnrs/lange/RNA_assemblies/a, /RNA_assemblies/) called at /opt/apps/trinity/2.2.0/Trinity line 2116 thread 2 main::prep_seqs('ARRAY(0x1211d38)', 'fq', 'right', 'R') called at /opt/apps/trinity/2.2.0/Trinity line 1317 thread 2 eval {...} called at /opt/apps/trinity/2.2.0/Trinity line 1317 thread 2 -conversion of 29941823 from FQ to FA format succeeded. Trinity run failed. Must investigate error above. Can I change something? Is this a bug? How to fix? Should I recreate this fastq file? I would like to know if this might be a software issue. Thanks, Iris

ADD COMMENT • link written 18 months ago by ilange • 0

Hi - The trinity problems have returned (along with a few cluster issues). The problem is not resolved yet. Please follow this other post for updates: https://biostar.usegalaxy.org/p/23175/

Thanks! Jen, Galaxy team

ADD REPLY • link written 18 months ago by Jennifer Hillman Jackson ♦ 25k

Use fastq input, not fasta. More above here: https://biostar.usegalaxy.org/p/22724/#24413

Also double check that the fastq datasets are not truncated (check the end of the uploaded datasets using the tool Text manipulation > Select last lines from a dataset (tail). 10 lines should be enough to see if the records are intact. The error message you got indicates a truncated input. This could have been due to network or other issues when the tool was having problems before, or it may be a problem with your input data.

ADD REPLY • link modified 15 months ago • written 15 months ago by Jennifer Hillman Jackson ♦ 25k

Please log in to add an answer.

Similar posts • Search »