ModelAngelo v1.0.1 stopped after predicting Cα

[yanjuisheng]

Hi jamaliki,

I would like to express my gratitude for your contributions to model building. However, I encountered an issue while using ModelAngelo to generate a protein model. After the Cα prediction step, ModelAngelo displayed an error message in the log file and halted the GNN execution. I have included the error message below. I am confident that the format of the fasta file containing three peptide sequences is correct, and there are no unknown residues in the file. I'm currently unable to determine the cause of the program's termination. Could you kindly provide any guidance on resolving this problem?

Thank you in advance.

2023-06-30 at 17:31:00 | ERROR | Error in ModelAngelo Traceback (most recent call last):

File "/opt/anaconda3/envs/model_angelo/bin/model_angelo", line 33, in sys.exit(load_entry_point('model-angelo==1.0.1', 'console_scripts', 'model_angelo')()) │ │ └ <function importlib_load_entry_point at 0x7fd1ce563d90> │ └ └ <module 'sys' (built-in)> File "/opt/anaconda3/envs/model_angelo/lib/python3.10/site-packages/model_angelo-1.0.1-py3.10.egg/model_angelo/main.py", line 52, in main args.func(args) │ │ └ Namespace(volume_path='/Users/XMi/Desktop/52/1/corner1.mrc', protein_fasta='/Users/XMi/Desktop/52/1/seq.fasta', rna_fasta=Non... │ └ <function main at 0x7fd1dafe8700> └ Namespace(volume_path='/Users/XMi/Desktop/52/1/corner1.mrc', protein_fasta='/Users/XMi/Desktop/52/1/seq.fasta', rna_fasta=Non...

File "/opt/anaconda3/envs/model_angelo/lib/python3.10/site-packages/model_angelo-1.0.1-py3.10.egg/model_angelo/apps/build.py", line 241, in main gnn_output = gnn_infer(gnn_infer_args) │ └ {'num_rounds': 3, 'crop_length': 200, 'repeat_per_residue': 1, 'esm_model': 'esm1b_t33_650M_UR50S', 'aggressive_pruning': Tru... └ <function infer at 0x7fd1daf1b400> File "/opt/anaconda3/envs/model_angelo/lib/python3.10/site-packages/model_angelo-1.0.1-py3.10.egg/model_angelo/gnn/inference.py", line 92, in infer protein = get_lm_embeddings_for_protein(lang_model, batch_converter, protein) │ │ │ └ Protein(atom_positions=None, atomc_positions=None, aatype=None, atom_mask=None, atomc_mask=None, residue_index=None, chain_in... │ │ └ <esm.data.BatchConverter object at 0x7fcf8e0e7fd0> │ └ ProteinBertModel( │ (embed_tokens): Embedding(33, 1280, padding_idx=1) │ (layers): ModuleList( │ (0-32): 33 x TransformerLa... └ <function get_lm_embeddings_for_protein at 0x7fd1daf1b520> File "/opt/anaconda3/envs/model_angelo/lib/python3.10/site-packages/model_angelo-1.0.1-py3.10.egg/model_angelo/data/generate_complete_prot_files.py", line 34, in get_lm_embeddings_for_protein protein_with_lm = add_lm_embeddings_to_protein(protein, lm_embeddings) │ │ └ array([[ 0.0707111 , 0.16950509, -0.00793343, ..., -0.3184964 , │ │ -0.15484953, -0.44611225], │ │ [ 0.24136926, -0.0... │ └ Protein(atom_positions=None, atomc_positions=None, aatype=None, atom_mask=None, atomc_mask=None, residue_index=None, chain_in... └ <function add_lm_embeddings_to_protein at 0x7fd1daf19cf0> File "/opt/anaconda3/envs/model_angelo/lib/python3.10/site-packages/model_angelo-1.0.1-py3.10.egg/model_angelo/utils/protein.py", line 897, in add_lm_embeddings_to_protein assert len(lm_embeddings) == input_protein.unified_seq_len │ │ └ 233 │ └ Protein(atom_positions=None, atomc_positions=None, aatype=None, atom_mask=None, atomc_mask=None, residue_index=None, chain_in... └ array([[ 0.0707111 , 0.16950509, -0.00793343, ..., -0.3184964 , -0.15484953, -0.44611225], [ 0.24136926, -0.0...

AssertionError: assert len(lm_embeddings) == input_protein.unified_seq_len

[jamaliki]

Hi,

Sorry for my late reply. This seems to be an issue in the input sequence. Have you made sure there are no non standard amino acids in your input? Could you share the sequence file?

Best, Kiarash.

[yanjuisheng]

Hi jamaliki,

Thank you for your help. I recreated a FASTA file and finally ModelAngelo proceeded.