when running humann2, stuck at bowtie2-build step

[zuota...@gmail.com]

Dear Humann2 developer,
When I use humann2, i got stuck at the the bowtie2-build step, looks like taking forever, and gave me error as below,
CRITICAL ERROR: Error executing: /users/b138776/bin/humann2/bin/bowtie2-build -f /lustre/project/SCNg/tao/projects/uc_mucosal_virome/humann2_output4/H06_mapped_humann2_temp_wh5irq/H06_mapped_custom_chocophlan_database.ffn /lustre/project/SCNg/tao/projects/uc_mucosal_virome/humann2_output4/H06_mapped_humann2_temp_wh5irq/H06_mapped_bowtie2_index --large-index

When I look at the .log, it shows,
10/09/2018 12:12:23 PM - humann2.utilities - CRITICAL: TRACEBACK:
Traceback (most recent call last):
File "/users/b138776/bin/humann2/lib/python2.7/site-packages/humann2/utilities.py", line 731, in execute_command
p = subprocess.check_call(cmd, stdin=stdin, stdout=stdout, stderr=stderr)
File "/users/b138776/bin/humann2/lib/python2.7/subprocess.py", line 190, in check_call
raise CalledProcessError(retcode, cmd)
CalledProcessError: Command '['/users/b138776/bin/humann2/bin/bowtie2-build', '-f', '/lustre/project/SCNg/tao/projects/uc_mucosal_virome/humann2_output4/H01_mapped_humann2_temp_Jgzo5O/H01_mapped_custom_chocophlan_database.ffn', '/lustre/project/SCNg/tao/projects/uc_mucosal_virome/humann2_output4/H01_mapped_humann2_temp_Jgzo5O/H01_mapped_bowtie2_index', '--large-index']' returned non-zero exit status 247

I tried use different bowtie versions and different humann2 installed by conda or pip installed humann2. The results were the same.
Would you pls kindly walk me out of this problem

[Eric Franzosa]

Can you clarify what "H01_mapped_custom_chocophlan_database.ffn" is? And specifically how many sequences it contains + total file size? It's possible that you're trying to index a very large database. I know for example when we index the entirety of ChocoPhlAn it is quite a long-running process.

Thanks,

Eric

[Ahmed Shibl]

Dear Eric,

I ran into a similar issue and got an identical error.

This is the command I used:

humann2 --resume --verbose --threads 4 --bypass-translated-search --bypass-prescreen --input ~/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined.fastq --output ~/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined_metaphlan_output --taxonomic-profile ~/miniconda3/envs/humann2/A3Bact_trial/A3Bact_combined_metaphlan_bugs_list.tsv --remove-temp-output

Please note that I used the --taxonomic-profile flag with the output from an independent metaphlan2 run (A3Bact_combined_metaphlan_bugs_list.tsv). (n.b. I also got the A3Bact_combined_metaphlan_bowtie2.txt  as output from the same metaphlan2 run but wasn't sure how to use it within the humann2 command.

This is the error I got at the bowtie2-build step (copy pasted from the log file):

01/10/2019 05:50:01 PM - humann2.search.nucleotide - INFO: Running bowtie2-build ........
01/10/2019 05:50:01 PM - humann2.utilities - DEBUG: Using software: /home/as11798/miniconda3/envs/humann2/bin/bowtie2-build
01/10/2019 05:50:01 PM - humann2.utilities - INFO: Execute command: /home/as11798/miniconda3/envs/humann2/bin/bowtie2-build -f /home/as11798/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined_metaphlan_output/Axenic_30mins_combined_humann2_temp_3P586z/Axenic_30mins_combined_custom_chocophlan_database.ffn /home/as11798/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined_metaphlan_output/Axenic_30mins_combined_humann2_temp_3P586z/Axenic_30mins_combined_bowtie2_index --large-index
01/12/2019 09:08:51 PM - humann2.utilities - CRITICAL: Error executing: /home/as11798/miniconda3/envs/humann2/bin/bowtie2-build -f /home/as11798/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined_metaphlan_output/Axenic_30mins_combined_humann2_temp_3P586z/Axenic_30mins_combined_custom_chocophlan_database.ffn /home/as11798/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined_metaphlan_output/Axenic_30mins_combined_humann2_temp_3P586z/Axenic_30mins_combined_bowtie2_index --large-index

01/12/2019 09:08:53 PM - humann2.utilities - CRITICAL: TRACEBACK:

Traceback (most recent call last):

  File "/home/as11798/miniconda3/envs/humann2/lib/python2.7/site-packages/humann2/utilities.py", line 731, in execute_command
    p = subprocess.check_call(cmd, stdin=stdin, stdout=stdout, stderr=stderr)
  File "/home/as11798/miniconda3/envs/humann2/lib/python2.7/subprocess.py", line 190, in check_call
    raise CalledProcessError(retcode, cmd)
CalledProcessError: Command '['/home/as11798/miniconda3/envs/humann2/bin/bowtie2-build', '-f', '/home/as11798/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined_metaphlan_output/Axenic_30mins_combined_humann2_temp_3P586z/Axenic_30mins_combined_custom_chocophlan_database.ffn', '/home/as11798/miniconda3/envs/humann2/BACMT_axenic_30min/Axenic_30mins_combined_metaphlan_output/Axenic_30mins_combined_humann2_temp_3P586z/Axenic_30mins_combined_bowtie2_index', '--large-index']' returned non-zero exit status 247

Any solutions to this? thanks!

Ahmed