openfold.data.feature_processing_multimer¶

Feature processing logic for multimer data pipeline.

Functions

`crop_chains`(chains_list, msa_crop_size, ...)	Crops the MSAs for a set of chains.
`pair_and_merge`(all_chain_features)	Runs processing on features to augment, pair and merge.
`process_final`(np_example)	Final processing steps in data pipeline, after merging and pairing.
`process_unmerged_features`(all_chain_features)	Postprocessing stage for per-chain features before merging.

crop_chains(chains_list, msa_crop_size, pair_msa_sequences, max_templates)¶

Crops the MSAs for a set of chains.

Parameters:

chains_list (List[Mapping[str, ndarray]]) – A list of chains to be cropped.
msa_crop_size (int) – The total number of sequences to crop from the MSA.
pair_msa_sequences (bool) – Whether we are operating in sequence-pairing mode.
max_templates (int) – The maximum templates to use per chain.

Returns:

The chains cropped.

Return type:

List[Mapping[str, ndarray]]

pair_and_merge(all_chain_features)¶

Runs processing on features to augment, pair and merge.

Parameters:: all_chain_features (MutableMapping[str, Mapping[str, ndarray]]) – A MutableMap of dictionaries of features for each chain.
Returns:: A dictionary of features.
Return type:: Mapping[str, ndarray]

process_final(np_example)¶

Final processing steps in data pipeline, after merging and pairing.

process_unmerged_features(all_chain_features)¶

Postprocessing stage for per-chain features before merging.

Parameters:: all_chain_features (MutableMapping[str, Mapping[str, ndarray]])