P

Pepmlm 650M

Developed by ChatterjeeLab
The first de novo generator of linear peptide binders dependent solely on target protein sequences
Downloads 396
Release Time : 9/27/2023

Model Overview

PepMLM is a target protein sequence-conditioned masked language modeling technique for generating peptide binders. It employs an innovative masking strategy to drive the ESM-2 protein language model to reconstruct binding regions, enabling the generation of candidate binders for any target protein without requiring structural information.

Model Features

No structural information required
Generates peptide binders relying solely on target protein sequences, without needing structural information of the target protein
Innovative masking strategy
Unique masking strategy that positions homologous peptide sequences at the terminus of the target protein sequence
Computational validation
In silico validation via AlphaFold-Multimer
Experimental validation
Achieved endogenous degradation of target substrates in cellular models for experimental validation

Model Capabilities

Peptide binder generation
Protein sequence analysis
Protein-peptide interaction prediction

Use Cases

Biomedical research
Targeted protein degradation
Generates peptides fused with E3 ubiquitin ligase domains for targeted degradation of specific proteins
Achieved endogenous degradation of target substrates in cellular models
Drug discovery
Generates candidate peptide binders for any target protein
Provides tools for downstream programmable proteome editing applications
Research tools
Protein interaction studies
Investigates mechanisms of protein-peptide interactions
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase