BiG-MAP.map.py Biom error

[AKBode]

Hi BiG-MAP developer

Thanks for developing the tool for the community.

I ended up with an error when running BiG-MAP.map for both metagenome and metatranscriptome data. One error example is listed down below. Any idea how to fix the error?

python3 $HOME/BiG-MAP/src/BiG-MAP.map.py -b $HOME/bigmap_input/mtx_S.txt -U /scratch1/users/MTX_filtered/*fastq.gz -O /scratch1/users/bigmap_output/results_mapping_MTX_hmg_ref_cf02_S/ -F $HOME/bigmap_output/results_family_genomes_cf02_mtx/ -th 64 -a True

Adding metadeta to biom and converting files into json format
Traceback (most recent call last):
File "/usr/users/BiG-MAP/src/BiG-MAP.map.py", line 1124, in
main()
File "/usr/users/BiG-MAP/src/BiG-MAP.map.py", line 1088, in main
biom_out1 = decoratebiom(biomfile, args.outdir, args.biom_output)
File "/usr/users/BiG-MAP/src/BiG-MAP.map.py", line 780, in decoratebiom
res_add = subprocess.check_output(cmd_sample, shell=True)
File "/usr/users/.conda/envs/BiG-MAP_process/lib/python3.6/subprocess.py", line 336, in check_output
**kwargs).stdout
File "/usr/users/.conda/envs/BiG-MAP_process/lib/python3.6/subprocess.py", line 418, in run
output=stdout, stderr=stderr)
subprocess.CalledProcessError: Command 'biom add-metadata -i /scratch1/users/bigmap_output/results_mapping_MTX_hmg_ref_cf02_S/BiG-MAP.map.biom -o /scratch1/users/bigmap_output/results_mapping_MTX_hmg_ref_cf02_S/BiG-MAP.map.meta.biom -m /usr/users/bigmap_input/mtx_S.txt --output-as-json' returned non-zero exit status 1

[victoriapascal]

Thanks for posting this here. Indeed, the BIOM step can be sometimes problematic. I would suggest to first check the metadata file is properly formatted: tab separated, ids matching the sample ids, first line starts with "#"... Unfortunately, we can not reproduce this error as this might be specific to your file/dataset. If the file is correctly formatted do you mind sharing the metadata file you are using with us?

[mhyleung]

Hi thanks for the response.

I made a silly mistake. I actually directed my metadata file to the wrong directory, and the resulting json file generated had no information. That's why it did not work for me. Sorry for the confusion.

I went back to changed the directory path and it worked. Thanks

[AKBode]

Thanks for posting this here. Indeed, the BIOM step can be sometimes problematic. I would suggest to first check the metadata file is properly formatted: tab separated, ids matching the sample ids, first line starts with "#"... Unfortunately, we can not reproduce this error as this might be specific to your file/dataset. If the file is correctly formatted do you mind sharing the metadata file you are using with us?

Hi Victoria,

I've double-checked the potential issues you mentioned. Things look fine and I still ended with the same error. Do you mind giving me your email? I'd share the files with you. Thank you.

[victoriapascal]

Sure. You can use the following email address: [email protected]

[AKBode]

I've resolved the issue by removing the flag "-a True"