`Could not get MSA/templates for ENSP00000359211_132T_S_543I_V: list index out of range`

Input file: an MSA created using colabfold with the reference sequence, and then mutating a single amino acid in the first sequence (1023P>S). Both sequences are exactly 1042 residues long.

Command and output:

[schilder@bamgpu16 ENST00000370192]$ colabfold_batch --save-single-representations --save-pair-representations af2_sameMSA/ af2_sameMSA/

WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1754084082.623474 2594776 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1754084082.760501 2594776 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1754084083.571507 2594776 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1754084083.571551 2594776 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1754084083.571552 2594776 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1754084083.571554 2594776 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2025-08-01 17:35:27,720 Running colabfold 1.5.5 (d1b8ec153b65a6b90e75a8d12f058ee292473962)
2025-08-01 17:35:54,118 non-fasta/a3m file in input directory: af2_sameMSA/log.txt
2025-08-01 17:35:56,290 Running on GPU
2025-08-01 17:35:59,805 Found 5 citations for tools or databases
2025-08-01 17:35:59,805 Query 1/162: ENSP00000359211_1023P_S (length 1025)
2025-08-01 17:36:36,992 Could not get MSA/templates for ENSP00000359211_1023P_S: list index out of range
Traceback (most recent call last):
  File "/grid/koo/home/schilder/projects/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 1286, in run
    = unserialize_msa(a3m_lines, query_sequence)
  File "/grid/koo/home/schilder/projects/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 926, in unserialize_msa
    seq = a3m_lines[i + 1]
IndexError: list index out of range
2025-08-01 17:36:37,031 Query 2/162: ENSP00000359211_1023P_T (length 1025)
2025-08-01 17:37:13,212 Could not get MSA/templates for ENSP00000359211_1023P_T: list index out of range
Traceback (most recent call last):
  File "/grid/koo/home/schilder/projects/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 1286, in run
    = unserialize_msa(a3m_lines, query_sequence)
  File "/grid/koo/home/schilder/projects/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 926, in unserialize_msa
    seq = a3m_lines[i + 1]
IndexError: list index out of range
2025-08-01 17:37:13,233 Query 3/162: ENSP00000359211_132T_S_543I_V (length 1025)
2025-08-01 17:37:46,905 Could not get MSA/templates for ENSP00000359211_132T_S_543I_V: list index out of range
Traceback (most recent call last):
  File "/grid/koo/home/schilder/projects/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 1286, in run
    = unserialize_msa(a3m_lines, query_sequence)
  File "/grid/koo/home/schilder/projects/localcolabfold/colabfold-conda/lib/python3.10/site-packages/colabfold/batch.py", line 926, in unserialize_msa
    seq = a3m_lines[i + 1]
IndexError: list index out of range
2025-08-01 17:37:46,926 Query 4/162: ENSP00000359211_142L_V (length 1025)
E0801 17:37:57.026062 2594776 ptx_compiler_helpers.cc:88] *** WARNING *** Invoking ptxas with version 12.3.52, which corresponds to a CUDA version <=12.6.2. CUDA versions 12.x.y up to and including 12.6.2 miscompile certain edge cases around clamping.
Please upgrade to CUDA 12.6.3 or newer.
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1754084280.982313 2594776 mlir_graph_optimization_pass.cc:425] MLIR V1 optimization pass is not enabled
2025-08-01 17:40:01,962 alphafold2_ptm_model_1_seed_000 recycle=0 pLDDT=80.6 pTM=0.75
2025-08-01 17:40:14,397 alphafold2_ptm_model_1_seed_000 recycle=1 pLDDT=84.8 pTM=0.825 tol=2.03
2025-08-01 17:40:26,853 alphafold2_ptm_model_1_seed_000 recycle=2 pLDDT=86.1 pTM=0.828 tol=7.76
2025-08-01 17:40:39,406 alphafold2_ptm_model_1_seed_000 recycle=3 pLDDT=90.1 pTM=0.885 tol=0.727
2025-08-01 17:40:39,407 alphafold2_ptm_model_1_seed_000 took 154.3s (3 recycles)
2025-08-01 17:40:52,480 alphafold2_ptm_model_2_seed_000 recycle=0 pLDDT=84.2 pTM=0.82
2025-08-01 17:41:04,984 alphafold2_ptm_model_2_seed_000 recycle=1 pLDDT=87 pTM=0.863 tol=1.29
2025-08-01 17:41:17,665 alphafold2_ptm_model_2_seed_000 recycle=2 pLDDT=91.1 pTM=0.899 tol=7.57
2025-08-01 17:41:30,387 alphafold2_ptm_model_2_seed_000 recycle=3 pLDDT=91.9 pTM=0.915 tol=0.429
2025-08-01 17:41:30,389 alphafold2_ptm_model_2_seed_000 took 50.3s (3 recycles)
2025-08-01 17:41:43,606 alphafold2_ptm_model_3_seed_000 recycle=0 pLDDT=88.6 pTM=0.887
2025-08-01 17:41:56,190 alphafold2_ptm_model_3_seed_000 recycle=1 pLDDT=87.9 pTM=0.859 tol=1.82
2025-08-01 17:42:09,040 alphafold2_ptm_model_3_seed_000 recycle=2 pLDDT=92.9 pTM=0.92 tol=6.86
2025-08-01 17:42:22,007 alphafold2_ptm_model_3_seed_000 recycle=3 pLDDT=93.8 pTM=0.928 tol=0.734
2025-08-01 17:42:22,007 alphafold2_ptm_model_3_seed_000 took 51.0s (3 recycles)
2025-08-01 17:42:35,121 alphafold2_ptm_model_4_seed_000 recycle=0 pLDDT=87.1 pTM=0.854
2025-08-01 17:42:47,769 alphafold2_ptm_model_4_seed_000 recycle=1 pLDDT=86.5 pTM=0.806 tol=3.57
2025-08-01 17:43:00,486 alphafold2_ptm_model_4_seed_000 recycle=2 pLDDT=92.2 pTM=0.904 tol=5.89
2025-08-01 17:43:13,319 alphafold2_ptm_model_4_seed_000 recycle=3 pLDDT=93.2 pTM=0.919 tol=0.41
2025-08-01 17:43:13,320 alphafold2_ptm_model_4_seed_000 took 50.7s (3 recycles)
2025-08-01 17:43:26,385 alphafold2_ptm_model_5_seed_000 recycle=0 pLDDT=83.2 pTM=0.819
2025-08-01 17:43:39,109 alphafold2_ptm_model_5_seed_000 recycle=1 pLDDT=91.1 pTM=0.91 tol=5.61
2025-08-01 17:43:51,911 alphafold2_ptm_model_5_seed_000 recycle=2 pLDDT=92.6 pTM=0.925 tol=0.554
2025-08-01 17:44:04,786 alphafold2_ptm_model_5_seed_000 recycle=3 pLDDT=93.2 pTM=0.928 tol=0.403
2025-08-01 17:44:04,787 alphafold2_ptm_model_5_seed_000 took 50.9s (3 recycles)
2025-08-01 17:44:05,385 reranking models by 'plddt' metric
2025-08-01 17:44:05,385 rank_001_alphafold2_ptm_model_3_seed_000 pLDDT=93.8 pTM=0.928
2025-08-01 17:44:05,386 rank_002_alphafold2_ptm_model_4_seed_000 pLDDT=93.2 pTM=0.919
2025-08-01 17:44:05,387 rank_003_alphafold2_ptm_model_5_seed_000 pLDDT=93.2 pTM=0.928
2025-08-01 17:44:05,388 rank_004_alphafold2_ptm_model_2_seed_000 pLDDT=91.9 pTM=0.915
2025-08-01 17:44:05,388 rank_005_alphafold2_ptm_model_1_seed_000 pLDDT=90.1 pTM=0.885

I've tried on both a A100 and an H100, both encountered the same error.
It seems to be specific to the MSA, though I can't figure out why.

Example data

Causes the error:
ENSP00000359211_1023P_S.a3m.txt

Doesn't cause the error:

ENSP00000359211_151N_D.a3m.txt

Start in colabfold/batch.py at unserialize_msa, especially the code around line 926 where the traceback indexes a3m_lines. Compare the provided failing and working .a3m files, then reproduce with the shown colabfold_batch command. Done means the failing MSA is handled without an IndexError and processing continues normally.