# Stable de novo protein design via joint conformational landscape and sequence optimization

## TL;DR

This paper introduces a new method for designing stable proteins by jointly optimizing their sequence and structure, leading to better folding stability.

## Contribution

The novel contribution is a joint optimization framework combining sequence-to-structure and structure-to-sequence modeling for improved de novo protein design.

## Key findings

- Joint optimization improves stability prediction and protein design compared to separate methods.
- Sequences from the joint model show more hydrophilic interactions, aiding structural stability.
- Large-scale experiments validate the effectiveness of the joint modeling approach.

## Abstract

Generative protein modeling provides advanced tools for designing diverse protein sequences and structures. However, accurately modeling the conformational landscape and designing sequences remain critical challenges: ensuring that the designed sequence reliably folds into the target structure as its most stable conformation, and optimizing the sequence for a given suboptimal fixed input structure. In this study, we present a systematic analysis of jointly optimizing sequence-to-structure and structure-to-sequence mappings. This approach enables us to find optimal solutions for modeling the conformational landscape. We validate our approach with large-scale protein stability measurements, demonstrating that joint optimization is superior for designing stable proteins using a joint model (TrRosetta and TrMRF) and for achieving high accuracy in stability prediction when jointly modeling (half-masked ESMFold pLDDT + ESM2 Pseudo-likelihood). We further investigate features of sequences generated from the joint model and find that they exhibit higher frequencies of hydrophilic interactions, which may help maintain both secondary structure registry and pairing-features not captured by structure-to-sequence modeling alone.

This study presents a comprehensive modelling framework that jointly optimizes sequence and structure to generate de novo proteins with improved folding stability, providing large-scale experimental benchmarking across multiple computational design methods