KosinskiLab / AlphaPulldown

https://doi.org/10.1093/bioinformatics/btac749
GNU General Public License v3.0
218 stars 49 forks source link

Some candidates in the prediction step lead to the following error : #427

Open ThisIsIt42 opened 1 month ago

ThisIsIt42 commented 1 month ago

Traceback (most recent call last): File "/opt/conda/bin/run_structure_prediction.py", line 274, in app.run(main) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 308, in run _run_main(main, args) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main sys.exit(main(argv)) File "/opt/conda/bin/run_structure_prediction.py", line 258, in main object_to_model, flags_dict, postprocess_flags, output_dir = pre_modelling_setup(interactors, FLAGS, output_dir = FLAGS.output_directory[index]) File "/opt/conda/bin/run_structure_prediction.py", line 184, in pre_modelling_setup object_to_model = MultimericObject( File "/opt/conda/lib/python3.10/site-packages/alphapulldown/objects.py", line 475, in init self.create_all_chain_features() File "/opt/conda/lib/python3.10/site-packages/alphapulldown/objects.py", line 660, in create_all_chain_features self.feature_dict = self.pair_and_merge( File "/opt/conda/lib/python3.10/site-packages/alphapulldown/objects.py", line 630, in pair_and_merge np_example = feature_processing.process_final(np_example) File "/opt/conda/lib/python3.10/site-packages/alphafold/data/feature_processing.py", line 167, in process_final np_example = _make_msa_mask(np_example) File "/opt/conda/lib/python3.10/site-packages/alphafold/data/feature_processing.py", line 191, in _make_msa_mask np_example['msa_mask'] *= seq_mask[None] ValueError: operands could not be broadcast together with shapes (4095,154) (1,200) (4095,154) Traceback (most recent call last): File "/opt/conda/bin/run_multimer_jobs.py", line 145, in app.run(main) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 308, in run _run_main(main, args) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main sys.exit(main(argv)) File "/opt/conda/bin/run_multimer_jobs.py", line 141, in main subprocess.run(" ".join(command), check=True, shell=True) File "/opt/conda/lib/python3.10/subprocess.py", line 526, in run raise CalledProcessError(retcode, process.args, subprocess.CalledProcessError: Command 'python3 /opt/conda/bin/run_structure_prediction.py --output_directory ./outputdir --num_cycle 3 --num_predictions_per_model 1 --data_directory /db --features_directory msas --pair_msa --nomsa_depth_scan --nomultimeric_template --fold_backend alphafold --nocompress_result_pickles --noremove_result_pickles --use_ap_style --use_gpu_relax --protein_delimiter + --models_to_relax None --input swrD+flgB' returned non-zero exit status 1.

DimaMolod commented 1 month ago

Hi @ThisIsIt42 , I think this may happen when you mix features generated by mmseqs2 and hmmer. Is it the case for you? If yes, please consider re-calculating all features either using mmseqs or hmmer. Please let me know if this helps :)

ThisIsIt42 commented 1 month ago

hello so I am just using mmseqs2 for the feature search. Its odd I have 109 candidates, 76 work but the other 33 seem to fail with the following error: Traceback (most recent call last): File "/opt/conda/bin/run_structure_prediction.py", line 274, in app.run(main) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 308, in run _run_main(main, args) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main sys.exit(main(argv)) File "/opt/conda/bin/run_structure_prediction.py", line 244, in main parsed_input = parse_fold(FLAGS.input, FLAGS.features_directory, FLAGS.protein_delimiter) File "/opt/conda/lib/python3.10/site-packages/alphapulldown/utils/modelling_setup.py", line 87, in parse_fold raise FileNotFoundError( FileNotFoundError: {'flgE'} not found in ['msas'] Traceback (most recent call last): File "/opt/conda/bin/run_multimer_jobs.py", line 145, in app.run(main) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 308, in run _run_main(main, args) File "/opt/conda/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main sys.exit(main(argv)) File "/opt/conda/bin/run_multimer_jobs.py", line 141, in main subprocess.run(" ".join(command), check=True, shell=True) File "/opt/conda/lib/python3.10/subprocess.py", line 526, in run raise CalledProcessError(retcode, process.args, subprocess.CalledProcessError: Command 'python3 /opt/conda/bin/run_structure_prediction.py --output_directory ./outputdir --num_cycle 3 --num_predictions_per_model 1 --data_directory /db --features_directory msas --pair_msa --nomsa_depth_scan --nomultimeric_template --fold_backend alphafold --nocompress_result_pickles --noremove_result_pickles --use_ap_style --use_gpu_relax --protein_delimiter + --models_to_relax None --input swrD+flgE' returned non-zero exit status 1.

This is just for the candidates that failed

DimaMolod commented 1 month ago

line 87, in parse_fold raise FileNotFoundError( FileNotFoundError: {'flgE'} not found in ['msas']

this line literally means that you don't have flgE.pkl file in your folder with features called msas. Could you check that you have all features files calculated for these 33 problematic candidates?

DimaMolod commented 1 week ago

Hi @ThisIsIt42 ! Just checking if you resolve your error and I can close this issue?