Models#

Models mapping biological sequences to feature representations.

Overview#

The seqme library provides a suite of models that convert biological sequences into numerical representations. These models fall into two main categories:

  • Embedding models — map sequences to fixed-length vector representations.

  • Property models — compute interpretable scalar or descriptor values such as charge, weight, or hydrophobicity.

Models#

seqme.models.ESM2

Wrapper for the ESM2 protein/peptide embedding model.

seqme.models.ESMFold

ESMFold protein language model.

seqme.models.ESMIF1

Wrapper for the ESM inverse folding (ESM-IF1) model.

seqme.models.Hyformer

Wrapper for the Hyformer molecule/peptide embedding model.

seqme.models.RNAFM

A language model trained on RNA sequences, which computes sequence-level embeddings by averaging token embeddings.

seqme.models.GENALM

GENA-LM is a family of Open-Source Foundational Models for Long DNA Sequences trained on human DNA sequence.

seqme.models.KmerFrequencyEmbedding

Computes normalized k-mer frequency embeddings for sequences.

seqme.models.AliphaticIndex

Aliphatic index of amino acid sequences.

seqme.models.Aromaticity

Aromaticity of amino acid sequences.

seqme.models.BomanIndex

Boman index, estimating binding potential to proteins.

seqme.models.Charge

Net charge of amino acid sequences at a given pH.

seqme.models.Gravy

GRAVY (hydropathy) score for amino acid sequences.

seqme.models.Hydrophobicity

Hydrophobicity using a selected scale.

seqme.models.HydrophobicMoment

Hydrophobic moment (i.e., amphiphilicity) for one or more amino acid sequences using a sliding-window approach.

seqme.models.InstabilityIndex

Instability index, predicting in vitro protein stability.

seqme.models.IsoelectricPoint

Isoelectric point of amino acid sequences.

seqme.models.ProteinWeight

Molecular weight of amino acid sequences.

seqme.models.LogP

Lipophilicity for SMILES sequences.

seqme.models.QED

Quantitative Estimate of Drug-likeness for SMILES sequences.

seqme.models.SAScore

Synthetic Accessibility Score (SA Score) for SMILES sequences.

Miscellaneous#

These models perform special-purpose operations or combine multiple models.

seqme.models.ThirdPartyModel

Wrapper for loading and calling a third-party plugin model.

seqme.models.PCA

Utility projecting sequence embeddings onto principle components.

seqme.models.Ensemble

Combines multiple predictor functions into a weighted ensemble.

Supported sequence types#

At-a-glance matrix of all models and supported sequence types.

✓ — supported, ✗ — not supported