ColabFold Relaxation Error Running Locally

Expected Behavior

I am unable to use --amber to relax models.

Current Behavior

First model will fold but crash immediately when the relaxation step begins.

Steps to Reproduce (for bugs)

conda update conda conda create -n colabfold python=3.7 conda activate colabfold pip install --upgrade pip pip install "colabfold[alphafold] @ git+https://github.com/sokrypton/ColabFold" conda install cudatoolkit conda install cudnn pip install "jax[cuda11_cudnn805]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html conda install -y -c conda-forge -c bioconda kalign2=2.04 hhsuite=3.3.0 conda install -y -c conda-forge openmm=7.5.1 pdbfixer (This does cause a small error that says failed with initial frozen solve but does complete, maybe this is where the problem lies) colabfold_batch 6CDX.fasta out1/ --num-models 1 --amber --use-gpu-relax

ColabFold Output (for bugs)

2022-06-27 20:23:01,839 Running colabfold 1.3.0 (2a47c6f1459fbbdb5242cbc62173f9b513813cfa) Downloading alphafold2 weights to /home/ubuntu/.cache/colabfold: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3.47G/3.47G [00:13<00:00, 277MB/s] WARNING: You are welcome to use the default MSA server, however keep in mind that it's a limited shared resource only capable of processing a few thousand MSAs per day. Please submit jobs only from a single IP address. We reserve the right to limit access to the server case-by-case when usage exceeds fair use.

If you require more MSAs:

You can precompute all MSAs with colabfold_search or
You can host your own API and pass it to --host-url 2022-06-27 20:23:16,563 generated new fontManager 2022-06-27 20:23:18,610 Found 6 citations for tools or databases 2022-06-27 20:23:22,591 Query 1/1: 6CDX (length 36) COMPLETE: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 150/150 [elapsed: 00:01 remaining: 00:00] 2022-06-27 20:23:23,763 Running model_3 2022-06-27 20:23:46,034 model_3 took 19.9s (3 recycles) with pLDDT 74.4 and ptmscore 0.432 Traceback (most recent call last): File "/home/ubuntu/.local/bin/colabfold_batch", line 8, in sys.exit(main()) File "/home/ubuntu/.local/lib/python3.8/site-packages/colabfold/batch.py", line 1724, in main run( File "/home/ubuntu/.local/lib/python3.8/site-packages/colabfold/batch.py", line 1369, in run outs, model_rank = predict_structure( File "/home/ubuntu/.local/lib/python3.8/site-packages/colabfold/batch.py", line 441, in predict_structure patch_openmm() File "/home/ubuntu/.local/lib/python3.8/site-packages/colabfold/batch.py", line 69, in patch_openmm from simtk.openmm import app ModuleNotFoundError: No module named 'simtk'

Your Environment

cudatoolkit-11.3.1
cudnn-8.2.1 gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0

Additionally, I at first thought my openmm install was incorrect, but running python -m simtk.testInstallation reports all differences within tolerance.

Jun 27 '22 20:06 RaymondRuff

Update, I have been able to get the code runner further by installing via yml (file attached as .txt). I can now us --amber but still unable to use --use-gpu-relax. When --use-gpu-relax is added the output is the error below.

2022-07-18 17:04:05,893 Running model_3 2022-07-18 17:04:30,079 model_3 took 21.3s (3 recycles) with pLDDT 76.8 and ptmscore 0.451 Traceback (most recent call last): File "/home/ubuntu/miniconda3/envs/colabfold1/bin/colabfold_batch", line 8, in sys.exit(main()) File "/home/ubuntu/miniconda3/envs/colabfold1/lib/python3.7/site-packages/colabfold/batch.py", line 1756, in main stop_at_score_below=args.stop_at_score_below, File "/home/ubuntu/miniconda3/envs/colabfold1/lib/python3.7/site-packages/colabfold/batch.py", line 1389, in run random_seed=random_seed, File "/home/ubuntu/miniconda3/envs/colabfold1/lib/python3.7/site-packages/colabfold/batch.py", line 479, in predict_structure relaxed_pdb_str, _, _ = amber_relaxer.process(prot=unrelaxed_protein) File "/home/ubuntu/miniconda3/envs/colabfold1/lib/python3.7/site-packages/alphafold/relax/relax.py", line 66, in process use_gpu=self._use_gpu) File "/home/ubuntu/miniconda3/envs/colabfold1/lib/python3.7/site-packages/alphafold/relax/amber_minimize.py", line 483, in run_pipeline use_gpu=use_gpu) File "/home/ubuntu/miniconda3/envs/colabfold1/lib/python3.7/site-packages/alphafold/relax/amber_minimize.py", line 419, in _run_one_iteration raise ValueError(f"Minimization failed after {max_attempts} attempts.") ValueError: Minimization failed after 100 attempts.

This is the same error as mentioned in issue 145 but no solution was ever specified beyond updating python which I have found causes other problems.

Any advice would be welcome! Thanks

environment.txt

Jul 18 '22 17:07 RaymondRuff

I use the localcolabfold installer, and it didn't reproduce the error with the latest commit of colabfold.

$ cat 6cdx.fasta
>6CDX_1
XGCILNGRTDLGTLLFRCRRDSDCPGACICRGNGYCG

$ colabfold_batch 6cdx.fasta out1/ --num-models 1 --amber --use-gpu-relax
2022-07-20 01:58:18,980 Running colabfold 1.3.0 (26de12d3afb5f85d49d0c7db1b9371f034388395)
WARNING: You are welcome to use the default MSA server, however keep in mind that it's a limited shared resource only capable of processing a few thousand MSAs per day. Please submit jobs only from a single IP address. We reserve the right to limit access to the server case-by-case when usage exceeds fair use.

If you require more MSAs:

* You can precompute all MSAs with `colabfold_search` or

* You can host your own API and pass it to `--host-url`
2022-07-20 01:58:20,173 Found 6 citations for tools or databases
2022-07-20 01:58:21,639 Query 1/1: 6CDX_1 (length 37)
COMPLETE: 100%|████████████████| 150/150 [elapsed: 00:01 remaining: 00:00]
2022-07-20 01:58:22,781 Running model_3
2022-07-20 01:58:37,940 model_3 took 13.6s (3 recycles) with pLDDT 71.8 and ptmscore 0.409
2022-07-20 01:58:47,105 Relaxation took 7.9s
2022-07-20 01:58:47,105 reranking models by plddt
2022-07-20 01:58:47,585 Done

conda install -y -c conda-forge openmm=7.5.1 pdbfixer (This does cause a small error that says failed with initial frozen solve but does complete, maybe this is where the problem lies)

The GPU relaxation depends on the openmm package. Could you share the error message?

Jul 19 '22 17:07 YoshitakaMo

Interestingly when I install using the localcolabfold installer I do not have an error and am able to use --use-gpu-relax. The error only occurs when following the colabfold documentation or the environment.yml I provided above. I will use the localcolabfold installer from now on to avoid the error.

Since you asked though, below is the full error I see when trying to use amber on gpu using the install from the environment.yml.

(colabfold) rruff@pplhpc1gn003:~$ colabfold_batch 6CDX.fasta out1/ --num-models 1 --amber --use-gpu-relax 2022-07-19 10:08:26,699 Running colabfold 1.3.0 (26de12d3afb5f85d49d0c7db1b9371f034388395) WARNING: You are welcome to use the default MSA server, however keep in mind that it's a limited shared resource only capable of processing a few thousand MSAs per day. Please submit jobs only from a single IP address. We reserve the right to limit access to the server case-by-case when usage exceeds fair use.

If you require more MSAs:

You can precompute all MSAs with colabfold_search or
You can host your own API and pass it to --host-url 2022-07-19 10:08:43,808 Found 6 citations for tools or databases 2022-07-19 10:08:48,139 Query 1/5: 6CDX1 (length 36) 2022-07-19 10:08:50,655 Running model_3 | 0/150 [elapsed: 00:00 remaining: ?] 2022-07-19 10:09:25,400 model_3 took 31.4s (3 recycles) with pLDDT 77.5 and ptmscore 0.456 Traceback (most recent call last): File "/home/rruff/miniconda3/envs/colabfold/bin/colabfold_batch", line 8, in sys.exit(main()) File "/home/rruff/miniconda3/envs/colabfold/lib/python3.7/site-packages/colabfold/batch.py", line 1756, in main stop_at_score_below=args.stop_at_score_below, File "/home/rruff/miniconda3/envs/colabfold/lib/python3.7/site-packages/colabfold/batch.py", line 1389, in run random_seed=random_seed, File "/home/rruff/miniconda3/envs/colabfold/lib/python3.7/site-packages/colabfold/batch.py", line 479, in predict_structure relaxed_pdb_str, _, _ = amber_relaxer.process(prot=unrelaxed_protein) File "/home/rruff/miniconda3/envs/colabfold/lib/python3.7/site-packages/alphafold/relax/relax.py", line 66, in process use_gpu=self._use_gpu) File "/home/rruff/miniconda3/envs/colabfold/lib/python3.7/site-packages/alphafold/relax/amber_minimize.py", line 483, in run_pipeline use_gpu=use_gpu) File "/home/rruff/miniconda3/envs/colabfold/lib/python3.7/site-packages/alphafold/relax/amber_minimize.py", line 419, in _run_one_iteration raise ValueError(f"Minimization failed after {max_attempts} attempts.") ValueError: Minimization failed after 100 attempts.

Jul 19 '22 18:07 RaymondRuff