OpenMed-NER-ProteinDetect-ModernClinical-395M

What it does

This model detects protein, protein complex, protein enumeration, protein family or group, and protein variant entities in English clinical notes and research papers. It reports an F1 score of 0.91 on the FSU-PRGE benchmark, using a ModernBERT backbone with 396M parameters and supporting sequences up to 8192 tokens.

Best for

• Protein mentions in clinical notes
• Protein variants in research papers
• Protein complexes in drug discovery

What it detects (5 labels)

• Protein
• Protein complex
• Protein (enumeration)
• Protein family or group
• Protein variant

Quality

Biomedical Entity Recognition - Various biomedical entities

Use it

pip install transformers torch

from transformers import pipeline
ner = pipeline("token-classification", model="OpenMed/OpenMed-NER-ProteinDetect-ModernClinical-395M", aggregation_strategy="simple")
print(ner("Patient prescribed 10 mg aspirin for hypertension."))

pip install --upgrade "openmed[onnx-runtime]"

from openmed import OnnxModel
model = OnnxModel.from_pretrained("OpenMed/OpenMed-NER-ProteinDetect-ModernClinical-395M-v1-onnx-android")
for entity in model("The patient started imatinib after the consultation."):
    print(entity.to_dict())