VIRSorter_global-phage-signal.csv has no content

[473021677]

When I was running VirSorter, it didn't report any error. But when the program finished, one result file, namely VIRSorter_global-phage-signal.csv hadn't any content. Can someone help me?

[jarrodscott]

Hi @473021677

I may be having a similar problem (see the later parts of this issue thread https://github.com/simroux/VirSorter/issues/28)

It seems like the issue can be traced back to Step_3. As suggested by the developers I ran /path/to/Virsorter/Scripts/Step_3_highlight_phage_signal.pl Metric_files/VIRSorter_affi-contigs.tab VIRSorter_phage_signal_test.csv 16

And got back a blank file. I am curious how you installed Virsorter because I suspect there is something wrong with my installation. I followed the recipe for the conda install

/path/to/Virsorter/Scripts/Step_3_highlight_phage_signal.pl Metric_files/VIRSorter_affi-contigs.tab VIRSorter_phage_signal_test.csv 16

[473021677]

Hi jarrodscott: Yes. I am sure that there must be something wrong with the installation because many dependency packages have been needed during the installation. I cann't solve the problem. So I have decided to use VirSorter in the iPlant discovery environment(https://de.iplantcollaborative.org/de/). And I have used the app to predict the viral genomes and prophages from my metagenomic datasets successfully.

Best, Yuan

[jarrodscott]

@473021677 Hi Yuan I was unable to solve the problem as well so I took your advice and have the datasets running on iPlant! Thanks for your help!

Best, Jarrod

[473021677]

Hi Jarrod It's my pleasure to help you.

Best, Yuan

------------------ Original ------------------ From: "Jarrod"[email protected]; Date: Wed, Jun 6, 2018 09:38 AM To: "simroux/VirSorter"[email protected]; Cc: "473021677"[email protected]; "Mention"[email protected]; Subject: Re: [simroux/VirSorter] VIRSorter_global-phage-signal.csv has nocontent (#30)

@473021677 Hi Yuan I was unable to solve the problem as well so I took your advice and have the datasets running on iPlant! Thanks for your help!

Best, Jarrod

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[simroux]

Hi Yuan and Jarrod,

I believe that in both your cases, the issue you had was with the C script used to calculate enrichment statistics. This should now be fixed, plus we have a new option ("--no_c") to tell VirSorter to not use this C script and instead use a perl alternative, which is slower but should work on most systems.

If you have the occasion, please re-install VirSorter the same way as you did before (i.e. follow the conda environment guide in the readme), and let me know if this solve your problems !

[istmobiome]

Hi Simon,

I reinstalled VirSorter (using the updated conda recipe) and ran the test fasta file you (or Bryan) sent. Looks like it worked great! I am attaching the err, log, and VIRSorter_global-phage-signal.csv files so you can have a look test_out.zip. Let me know if anything looks funny.

If all looks good this can close my original issue which started off as a Diamond version problem #28

Thanks to you and Bryan for all the help and an awesome program.

[istmobiome]

oops, here is the command I ran

wrapper_phage_contigs_sorter_iPlant.pl -f input_sequences.fna --ncpu 20 --db 2 --wdir test_master --data-dir ~/virsorter-data --diamond and it looks like the default for --no_c is 0. So it ran fine with the C script?

[simroux]

Hi Jarrod,

Yes, the default is still to use the C script, it has been updated so that it should now compile and run on more modern gcc, so I was hopeful your problem would have been fixed (and it seems like it was !). The "no_c" option has been added mostly to avoid future cases like this, i.e. if in the future there is a new issue with the C script, users will be able to use "no_c" to get a VirSorter result.

[jarrodscott]

roger that. Awesome to have this working. thanks again!

On Sun, Jun 10, 2018 at 3:19 PM simroux [email protected] wrote:

Hi Jarrod,

Yes, the default is still to use the C script, it has been updated so that it should now compile and run on more modern gcc, so I was hopeful your problem would have been fixed (and it seems like it was !). The "no_c" option has been added mostly to avoid future cases like this, i.e. if in the future there is a new issue with the C script, users will be able to use "no_c" to get a VirSorter result.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/simroux/VirSorter/issues/30#issuecomment-396079020, or mute the thread https://github.com/notifications/unsubscribe-auth/APcSthDuBs6HDC5G-JUkggT2F89ISAYcks5t7X9YgaJpZM4UQHKe .

[Sapsanas]

Hi Simon,

I am very sorry for bothering you again, but as I see, this update should have solved the issue that I faced as well https://github.com/simroux/VirSorter/issues/29 However, after reinstalling VirSorter using conda and specifying the option --no_c, I still do not get any results. Could you please help me with?

Please find attached the file with output after running updated VirSorter.

VirSorter_Sapsanas.zip

Thank you in advance.

[simroux]

Hi,

Based on the log file, it looks like you have been trying to run VirSorter multiple times on the same directory. Could you try again with the same input file but a new working folder ?

Thanks,

[Sapsanas]

Dear Simon,

I rerun VirSorter on this sample, please find attached the output. VirSorter_Sapsanas_new.zip

[simroux]

Hi,

It looks like metagene_annotator was not installed, so VirSorter could not predict proteins. Could you double check on the README that you followed the specific step after "Finally, you'll need to download MetaGeneAnnotator ... " ?

[Sapsanas]

Dear Simon,

thank you for your response! Indeed, I downloaded Metagen_Annotator to new folder bin/metagene because of instruction (tar -xvzf metagene/mga_x86_64.tar.gz), transferring the mga_linux_ia64 to /bin/ helped! Thank you again, I think that issue #29 could be closed.

[thierryjanssens]

Dear Simon,

I experience the same problem, i.e. an empty final csv file, despite the prediction o significant prophages by other packages. mga_linux_ia64 has been installed in the env bin folder. Also the option --no_c does not relieve this problem. I am looking forward for your help.

Kind regards,

Thierry

[simroux]

Dear Thierry,

Could you send the log files of the VirSorter run ? There are (unfortunately) a number of reasons why the final csv would be empty, but hopefully we can figure what's going on there.

Thanks, Best, Simon

[thierryjanssens]

Hello Simon,

have you received the log files in the email of 2 days ago?

Regards,

Thierry

[simroux]

Hi Thierry,

Sorry, I didn't receive any email. You can attach them to your messages here, or send them to sroux (at) lbl.gov