Some models are not available through e.g., PyPI or Huggingface - only the git repository may be available. Other models licenses may not be compatible with seqme's license. Hence, why these models are not directly in the seqme package. Here we provide a repository for setting up such a third-party model.
| Model | Repository | Description |
|---|---|---|
| AMPlify | seqme-amplify | Attentive deep learning model for antimicrobial peptide (AMP) prediction |
| amPEPpy | seqme-amPEPpy | Random forest classifier for antimicrobial peptide prediction using global protein sequence descriptors |
| ESM-IF1 | seqme-esmif1 | Inverse folding model that generates amino acid sequences from fixed 3D protein backbones |
An external model is compatible with seqme if it is setup using uv (lockfile, python version defined), and defines an entry point (function).
Setup a project using:
uv init --package hello-modelRun the model:
import seqme as sm
model = sm.models.ThirdPartyModel(
entry_point="hello_model.model:embed",
path="../thirdparty/hello-model",
url="https://github.com/szczurek-lab/seqme-thirdparty",
branch="main",
)
model(sequences=["SEQVENCE"])