AlphaFold2_batch, MSA single_sequence, use_templates

I am trying to run colabfold batch with single_sequence MSA mode and use_templates=True.

Expected Behavior

Finish colabfold batch and get result.

Current Behavior

After running one run, following error occurs:

2024-04-29 13:37:05,798 Running on GPU
2024-04-29 13:37:08,920 Found 5 citations for tools or databases
2024-04-29 13:37:08,921 Query 1/5: 101 (length 219)
2024-04-29 13:37:46,312 Sequence 0 found templates: ['6aru_A', '8hgs_A', '3njp_B', '7sz5_A', '2ahx_B', '7sz5_B', '1yy9_A', '2ahx_A', '3u7u_C', '3u7u_A', '3u2p_A', '5wb8_D', '5wb7_A', '7len_A', '1mox_A', '7lfs_A', '3u9u_F', '5wb8_A', '4uv7_A', '4kro_A']
2024-04-29 13:37:46,314 Sequence 1 found no templates
2024-04-29 13:38:34,916 alphafold2_multimer_v3_model_1_seed_000 recycle=0 pLDDT=76.7 pTM=0.687 ipTM=0.176
2024-04-29 13:38:40,045 alphafold2_multimer_v3_model_1_seed_000 recycle=1 pLDDT=89.8 pTM=0.892 ipTM=0.845 tol=2.4
2024-04-29 13:38:45,076 alphafold2_multimer_v3_model_1_seed_000 recycle=2 pLDDT=88.8 pTM=0.876 ipTM=0.799 tol=0.231
2024-04-29 13:38:45,077 alphafold2_multimer_v3_model_1_seed_000 took 43.3s (2 recycles)
---------------------------------------------------------------------------
UFuncTypeError                            Traceback (most recent call last)
[<ipython-input-4-df19c07a99ab>](https://localhost:8080/#) in <cell line: 17>()
     15 
     16 queries, is_complex = get_queries(input_dir)
---> 17 run(
     18     queries=queries,
     19     result_dir=result_dir,

1 frames
[/content/colabfold/batch.py](https://localhost:8080/#) in predict_structure(prefix, result_dir, feature_dict, is_complex, use_templates, sequences_lengths, pad_len, model_type, model_runner_and_params, num_relax, relax_max_iterations, relax_tolerance, relax_stiffness, relax_max_outer_iterations, rank_by, random_seed, num_seeds, stop_at_score, prediction_callback, use_gpu_relax, save_all, save_single_representations, save_pair_representations, save_recycles)
    495 
    496             # early stop criteria fulfilled
--> 497             if mean_scores[-1] > stop_at_score: break
    498 
    499         # early stop criteria fulfilled

UFuncTypeError: ufunc 'greater' did not contain a loop with signature matching types (<class 'numpy.dtypes.Float16DType'>, <class 'numpy.dtypes.StrDType'>) -> None

Your Environment

Running on Google Colab,
T4 GPU, 15GB

Reproduce the batch run on Google Colab using single_sequence MSA mode with use_templates=True. Start in batch.py at predict_structure, especially the comparison at line 497, and inspect the values involved in the reported NumPy error. Done means the run completes without the UFuncTypeError and produces its result.