mPARCE

mPARCE performs iterative computational evolution of peptide sequences to optimize protein-peptide binding affinities using Rosetta backrub conformational sampling and a consensus of protein-ligand scoring functions.

Key Features:

• Iterative Computational Evolution Algorithm: Implements an iterative algorithm inspired by the PARCE protocol that performs single-point mutations to explore peptide sequence space.
• Integration with Rosetta Framework: Leverages Rosetta modules, specifically backrub sampling, for conformational sampling of protein-peptide complexes.
• Consensus Metric for Scoring: Applies a consensus metric combining multiple open-source protein-ligand scoring functions to estimate binding affinities at each mutation step.
• Guided Mutations: Allows mutations to be guided by structural properties or prior biological knowledge for targeted optimization.
• Iterative Optimization Process: Evaluates peptide modifications at each iteration based on score differences from the consensus metric to iteratively refine sequences.
• Benchmarking and Validation: Sampling and scoring scheme has been benchmarked against protein-peptide complexes with known experimental affinity values.
• Customization and Flexibility: Provides code-level customization to adapt computational settings to specific applications.

Scientific Applications:

• Drug discovery and development: Supports optimization of peptide therapeutics to improve efficacy and stability.
• Peptide design with NNAAs: Facilitates design of peptides incorporating non-natural amino acids (NNAAs) to enhance binding affinity and stability.

Methodology:

Starts from an initial peptide sequence, performs iterative single-point mutations guided by structural properties or biological knowledge, applies Rosetta backrub conformational sampling, and scores variants using a consensus of multiple protein-ligand scoring functions.