multimer model error:config_dict.py", line 903, in __getitem__ field = self._fields[key] KeyError: 'data'

[liuqs1990]

Expected Behavior

could predict the protein complex structure

Current Behavior

error occurred.

Steps to Reproduce (for bugs)

I am following the localcolabfold and I am able to predict a single protein. Working well: colabfold_batch \ --amber \ --templates \ --num-recycle 1 \ --model-type AlphaFold2-ptm \ /home/qiushi/RESC6/T1083.fasta \ /home/qiushi/RESC6/T1083outdirredo

I also tried T1084 and T1085 (two proteins in one .fasta file as input, in two individual .fasta files as input and in a .csv file as input) and run:

colabfold_batch \ > --amber \ > --templates \ > --num-recycle 1 \ > --model-type AlphaFold2-multimer \ > /home/qiushi/RESC6/complex.fasta \ > /home/qiushi/RESC6/complexdir Then the error happened.

"/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/ml_collections/config_dict/config_dict.py", line 903, in getitem field = self._fields[key] KeyError: 'data'

ColabFold Output (for bugs)

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, please host your own API and pass it to --host-url 2022-03-07 17:23:01,398 Running colabfold 1.2.0 (46ae46946d771af1745d55c2c40050e8f54f5e17) 2022-03-07 17:23:01,400 Found 8 citations for tools or databases 2022-03-07 17:23:01.584816: E external/org_tensorflow/tensorflow/stream_executor/cuda/cuda_driver.cc:771] failed to alloc 17179213824 bytes unified memory; result: CUDA_ERROR_OUT_OF_MEMORY: out of memory 2022-03-07 17:26:22,830 Query 1/2: T1084 (length 73) COMPLETE: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 150/150 [elapsed: 00:03 remaining: 00:00]2022-03-07 17:26:26,196 Sequence 0 found no templates 2022-03-07 17:26:26,216 Running model_3 Traceback (most recent call last): File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/ml_collections/config_dict/config_dict.py", line 903, in getitem field = self._fields[key] KeyError: 'data' During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/ml_collections/config_dict/config_dict.py", line 827, in getattr return self[attribute] File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/ml_collections/config_dict/config_dict.py", line 909, in getitem raise KeyError(self._generate_did_you_mean_message(key, str(e))) KeyError: "'data'" During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/bin/colabfold_batch", line 8, in sys.exit(main()) File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/colabfold/batch.py", line 1386, in main save_pair_representations=args.save_pair_representations, File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/colabfold/batch.py", line 1102, in run prediction_callback=prediction_callback, File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/colabfold/batch.py", line 198, in predict_structure use_templates, File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/colabfold/batch.py", line 132, in batch_input eval_cfg = model_config.data.eval File "/home/qiushi/localcolabfold/colabfold_batch/colabfold-conda/lib/python3.7/site-packages/ml_collections/config_dict/config_dict.py", line 829, in getattr raise AttributeError(e) AttributeError: "'data'"

Your Environment

nvcc --version CUDA 11.5

any suggestions would be appreciated.thanks.

[bayarovici]

What worries me is that you are getting CUDA_ERROR_OUT_OF_MEMORY suggesting your protein is too long for the capability of your graphics card. I used to get keyerror until I made sure that my fasta/csv file is correctly formatted. I would start with the minimal possible example. Create a csv file in notepad with the following rows: id,sequence simple,GPLSVPEGAIASLNCTY

Save it and run it. But I'm not sure it will work in your case!