Hi,

I'm trying to help someone run the RNA-seq pipeline and they ran into this error after the TopHat step that says the assemblies could not be read (Screenshots at 10_26_33 and 10_26_14).  The file does exist and seems to have no issues with format, so I'm not sure why this error would occur.  Any ideas?  We're using UGENE 1.21 on Ubuntu Linux.

I also noticed a warning by Cufflinks (screenshot with "cufflinks" in the name) that says it can't connect to the update server to verify the current version.  It's possible that they have not changed the location of the update server in this version of Cufflinks, though their website has changed.  It usually keeps checking for about 5 minutes, so when we're testing the pipeline with small datasets that otherwise should run quickly, this is a long time to wait unnecessarily.  I would suggest adding the --no-update-check command to cufflinks (and cuffdiff) commands to skip this step.

Thanks,

Andrew

---

Hello! Thanks for the detailed report and sorry for the late reply!

Regarding the first error, it seems that TopHat was not able to map the reads, so there were no output. Please do the following:
1) Make sure that the reference sequence, that you provided, is the correct one for the data.
2) Try to filter/trim/etc. your input NGS data (i.e. FASTQ files) BEFORE using the Tuxedo workflow. For example, use this workflow: https://ugene.unipro.ru/wiki/display/WDD21/Raw+RNA-Seq+Data+Processing.
3) Check out any errors of TopHat in the log.

Also, see these NGS tutorials: LINK. There are some sample data, you may run the workflow using them to make sure that it works in your environment.

As for the second error, we will follow your advice and fix this issue in the nearest future (see issue UGENE-5255). And we will also correct the error message in the first case (UGENE-5256). Thanks again for the report!

Hi Olga,

Thanks for the reply.  That's great you can fix the update-check issue for Cufflinks tools.

For the first issue, what I meant to say is that the output accepted_hits.bam file does exist, so TopHat appears to have mapped the reads successfully.  I have inspected the BAM file and the format is correct and I also inspected the TopHat logs and I don't see any issue with the TopHat job, which is why I suspected that the problem might be with UGENE. One potential clue might be in the fact that the UGENE element that is reporting the error is "Find Splice Junctions [with TopHat]", even though the error states that the BAM file can't be read.  Why would this element need to read the BAM file?  The BAM file is an output file of this element.

Thanks,

Andrew

Hi Andrew!


Quote:

Why would this element need to read the BAM file?

The element passes this file to the next element - Cufflinks in this case.

Is it possible, that you send us the BAM file and (ideally) the input data also? We will try to reproduce the issue and investigate the problem with this BAM.

Are you able to open the assembled data in UGENE?

Hello, Andrew.

There must be an error message in the UGNE log. It can tell about the reasons of the problem. Could you send it to us?
1) Click the button "3: Log" in the bottom of the UGENE window or just press Alt+3.
2) Select all content of the opened bar and copy it.

I suppose it can help us to find the problem reason.

---