seqme.models.ESMFold

seqme.models.ESMFold#

class seqme.models.ESMFold(*, device=None, batch_size=256, cache_dir=None, verbose=False)[source]#

ESMFold protein language model.

The model predicts the 3D-structure (fold) of a protein sequence.

Installation: pip install "seqme[esmfold]"

Reference:

Lin et al., “Language models of protein sequences at the scale of evolution enable accurate structure prediction” (https://www.biorxiv.org/content/10.1101/2022.07.20.500902v3)

__init__(*, device=None, batch_size=256, cache_dir=None, verbose=False)[source]#

Initialize the ESMFold model.

Parameters:

  • device (Optional[str]) – Device to run inference on, e.g., "cuda" or "cpu".

  • batch_size (int) – Number of sequences to process per batch.

  • cache_dir (Optional[str]) – Directory to cache the model.

  • verbose (bool) – Whether to display a progress bar.

__call__(sequences)[source]#

Call self as a function.

Return type:

list[ndarray]

Methods

__init__(*[, device, batch_size, cache_dir, ...])

Initialize the ESMFold model.

__call__(sequences)

Call self as a function.

fold(sequences, *[, convention, ...])

Predict protein sequences TM-score, pLDDT and 3D-structure, i.e., atom coordinates.
Contents