No PROTS_FIRST_ROUND.csv

[tthye1]

Hello,

When running the docker version of the viga script I got the following errror message.
Running DIAMOND to predict the protein function according to homology inference using default parameters
Traceback (most recent call last):
File "/program/VIGA.py", line 713, in
with open("PROTS_FIRST_ROUND.csv", "rU") as diamondresults:
IOError: [Errno 2] No such file or directory: 'PROTS_FIRST_ROUND.csv'

Somehow diamond does not create the .csv file.
What went wrong?
Best,

Thorsten

[EGTortuero]

Hi Thorsten,

First of all, thank you for using VIGA to annotate viral sequences. I did not check yet the docker version but, according to the source code, in principle there would not be problems as PROTS_FIRST_ROUND.csv should be created after running diamond (or creating as an empty file in case there were no proteins at all). I have a simple question in order to detect the real matter:

1. Did you try to run diamond out of VIGA with the following command:
   diamond blastp -q PROTS_FIRST_ROUND.faa -d <reference_database> -e 0.00001 -f 6 -o PROTS_FIRST_ROUND.csv -p <number_cpus>?
   If yes, may I kindly ask for the output of the program? That could help to improve the pipeline

Thanks in advance,

Enrique

[tthye1]

Hi Enrique, Thanks for your quick response. I run the “diamond blastp -q PROTS_FIRST_ROUND.faa -d /viga-master/data/databases/RefSeq_Viral_DIAMOND/refseq_viral_proteins.dmnd -e 0.00001 -f 6 -o PROTS_FIRST_ROUND.csv -p 11” which generated the .csv file without errors. I used for this analysis a very reduced data set just to get the pipeline going. When I use the docker version I again get the error. Best wishes, Thorsten #CPU threads: 11 Scoring parameters: (Matrix=BLOSUM62 Lambda=0.267 K=0.041 Penalties=11/1) Temporary directory: Opening the database... [2.9e-05s] #Target sequences to report alignments for: 25 Opening the input file... [1.6e-05s] Opening the output file... [1.6e-05s] Loading query sequences... [3.7e-05s] Masking queries... [0.001884s] Building query seed set... [0.001109s] Algorithm: Query-indexed Building query histograms... [0.000798s] Allocating buffers... [3e-05s] Loading reference sequences... [0.173565s] Building reference histograms... [0.115761s] Allocating buffers... [3.1e-05s] Initializing temporary storage... [0.002589s] Processing query chunk 0, reference chunk 0, shape 0, index chunk 0. Building reference index... [0.062766s] Building query index... [0.001759s] Building seed filter... [0.000853s] Searching alignments... [0.005462s] Deallocating buffers... [5e-06s] Computing alignments... [0.036181s] Deallocating reference... [0.004915s] Loading reference sequences... [1.3e-05s] Deallocating buffers... [0.000118s] Deallocating queries... [4e-06s] Loading query sequences... [1e-05s] Closing the input file... [5e-06s] Closing the output file... [3.3e-05s] Closing the database file... [3e-06s] Deallocating taxonomy... [3e-06s] Total time = 0.408486s Reported 72 pairwise alignments, 72 HSPs. 5 queries aligned.

…

On 4. Jun 2019, at 12:55, Enrique Gonzalez-Tortuero ***@***.***> wrote: Hi Thorsten, First of all, thank you for using VIGA to annotate viral sequences. I did not check yet the docker version but, according to the source code, in principle there would not be problems as PROTS_FIRST_ROUND.csv should be created after running diamond (or creating as an empty file in case there were no proteins at all). I have a simple question in order to detect the real matter: Did you try to run diamond out of VIGA with the following command: diamond blastp -q PROTS_FIRST_ROUND.faa -d <reference_database> -e 0.00001 -f 6 -o PROTS_FIRST_ROUND.csv -p <number_cpus>? If yes, may I kindly ask for the output of the program? That could help to improve the pipeline Thanks in advance, Enrique — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#6?email_source=notifications&email_token=AHWFHMWDOSO56F5FYHGWAVTPYZC2RA5CNFSM4HS2ANW2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODW4GD5I#issuecomment-498622965>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AHWFHMV453DPBSOA25F7FHLPYZC2RANCNFSM4HS2ANWQ>.

-- Dr. Thorsten Thye Bernhard Nocht Institute for Tropical Medicine Infectious Disease Epidemiology Bernhard Nocht Str. 74 20359 Hamburg Germany phone +49 40 42818-403 @ thye@bni-hamburg.de <mailto:c.g.meyer@bni.uni-hamburg.de>

[EGTortuero]

Thanks for your input.

That is very odd. In principle I would expect to see such behaviour if Diamond did not create the file at all regardless the docker. I will try to replicate your error in the docker version and I hope to bring a solution for this issue asap.

Best,

Enrique

[FTouzain]

Update 2022, error message is:
Running DIAMOND to predict the protein function according to homology inference using default parameters
Error: Database was built with a different version of Diamond and is incompatible.
Traceback (most recent call last):
File "/program/VIGA.py", line 713, in
with open("PROTS_FIRST_ROUND.csv", "rU") as diamondresults:
IOError: [Errno 2] No such file or directory: 'PROTS_FIRST_ROUND.csv'

Which Diamond version do you use in docker please? (because we need to install Diamond to create databases, but indexes created by last version are incompatibles with the one you use).

Thank you.
Best regards.