RFOptimization: Guiding Design Optimization with All-Atom Structure Prediction
- 1University of Washington
Abstract
Biomolecular interactions, including protein--protein interactions, protein--nucleic acid recognition, and protein--small molecule binding, underlie a wide range of biological processes and therapeutic mechanisms. Although recent \emph{de novo} design methods can generate candidate binders for diverse molecular targets, practical design campaigns remain limited by low filter-passing rates and model-specific biases that arise when designs are optimized against a single predictor. Here, we present RFOptimization (RFO), a training-free framework for all-atom biomolecular binder optimization. RFO formulates binder improvement as a residue-wise mutational search problem, sampling candidate substitutions alternately based on gradient-guided sequence optimization using all-atom structure prediction models and a cycling-based sequence redesign strategy that alternates structure generation with an orthogonal predictor and MPNN-based sequence design to improve the \emph{in silico} success rate of RFdiffusion-generated binders within minutes of computation. To reduce overfitting to any individual structure model, candidate mutations are further evaluated with orthogonal AlphaFold3 metrics as final filters. We demonstrate the generality of RFO across diverse design settings, including classical protein binder design, ligand-binding biosensor design, cyclic peptide design, and active site-aware enzyme design.
Related
- Probe Before You Edit: Probing-Guided Molecular Optimization for LLM Agents in Structure-Based Drug Design
- What Transfers Across Protein Pockets? Experience Reuse in Molecular Optimization for Agentic Structure-Based Drug Design
- 1017 Computational design of de novo antigen-specific T cell receptors using AI optimization and long-time-scale molecular simulations for targeted cancer immunotherapy
- Efficient Microwave-Enhanced Solid Phase Peptide Synthesis of Secretin Hormones Amenable for Rapid Library Design and Lead Optimization
- KAT-Net: a hyperbolic routing framework for joint lysine acylation site prediction with shared sequence–structure representations
BibTeX
@misc{zhang2026rfoptimization,
title = {{RFOptimization: Guiding Design Optimization with All-Atom Structure Prediction}},
author = {O. Zhang and J. Wang and T. R. Thompson and Z. You and Z. Song and F. DiMaio and D. Baker},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.04.749184},
url = {https://www.biorxiv.org/content/10.64898/2026.09.04.749184v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - RFOptimization: Guiding Design Optimization with All-Atom Structure Prediction
AU - O. Zhang
AU - J. Wang
AU - T. R. Thompson
AU - Z. You
AU - Z. Song
AU - F. DiMaio
AU - D. Baker
PY - 2026
DA - 2026/09/04
PB - bioRxiv
DO - 10.64898/2026.09.04.749184
UR - https://www.biorxiv.org/content/10.64898/2026.09.04.749184v1
AB - Biomolecular interactions, including protein--protein interactions, protein--nucleic acid recognition, and protein--small molecule binding, underlie a wide range of biological processes and therapeutic mechanisms. Although recent \emph{de novo} design methods can generate candidate binders for diverse molecular targets, practical design campaigns remain limited by low filter-passing rates and model-specific biases that arise when designs are optimized against a single predictor. Here, we present RFOptimization (RFO), a training-free framework for all-atom biomolecular binder optimization. RFO formulates binder improvement as a residue-wise mutational search problem, sampling candidate substitutions alternately based on gradient-guided sequence optimization using all-atom structure prediction models and a cycling-based sequence redesign strategy that alternates structure generation with an orthogonal predictor and MPNN-based sequence design to improve the \emph{in silico} success rate of RFdiffusion-generated binders within minutes of computation. To reduce overfitting to any individual structure model, candidate mutations are further evaluated with orthogonal AlphaFold3 metrics as final filters. We demonstrate the generality of RFO across diverse design settings, including classical protein binder design, ligand-binding biosensor design, cyclic peptide design, and active site-aware enzyme design.
ER -