# Boltzmann-Aligned Inverse Folding Model as a Predictor of Mutational Effects on Protein-Protein Interactions

arXiv: https://arxiv.org/abs/2410.09543 · HTML: https://mlsanity.com/paper/2410.09543/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 49 references that are in the map.
- First posted on arXiv 12 Oct 2024.

## Builds on (most cited first, 19 of 19)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [A General Theoretical Paradigm to Understand Learning from Human Preferences](https://mlsanity.com/paper/2310.12036/) (2023)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [AlphaFold Meets Flow Matching for Generating Protein Ensembles](https://mlsanity.com/paper/2402.04845/) (2024)
- [Tranception: protein fitness prediction with autoregressive transformers and inference-time retrieval](https://mlsanity.com/paper/2205.13760/) (2022)
- [EigenFold: Generative Protein Structure Prediction with Diffusion Models](https://mlsanity.com/paper/2304.02198/) (2023)
- [Direct Preference-based Policy Optimization without Reward Modeling](https://mlsanity.com/paper/2301.12842/) (2023)
- [End-to-End Learning on 3D Protein Structure for Interface Prediction](https://mlsanity.com/paper/1807.01297/) (2018)
- [DiffDock-PP: Rigid Protein-Protein Docking with Diffusion Models](https://mlsanity.com/paper/2304.03889/) (2023)
- [Learning to Predict Mutation Effects of Protein-Protein Interactions by Microenvironment-aware Hierarchical Prompt Learning](https://mlsanity.com/paper/2405.10348/) (2024)
- [Predicting mutational effects on protein-protein binding via a side-chain diffusion probabilistic model](https://mlsanity.com/paper/2310.19849/) (2023)
- [Preference Optimization for Molecular Language Models](https://mlsanity.com/paper/2310.12304/) (2023)
- [Aligning Transformers with Continuous Feedback via Energy Rank Alignment](https://mlsanity.com/paper/2405.12961/) (2024)
- [Preference Optimization for Molecule Synthesis with Conditional Residual Energy-based Models](https://mlsanity.com/paper/2406.02066/) (2024)
- [Multi-level Interaction Modeling for Protein Mutational Effect Prediction](https://mlsanity.com/paper/2405.17802/) (2024)
- [Preference optimization of protein language models as a multi-objective binder design paradigm](https://mlsanity.com/paper/2403.04187/) (2024)

## Built on by (most cited first, 5 of 5)

- [ProteinZero: Self-Improving Protein Generation via Online Reinforcement Learning](https://mlsanity.com/paper/2506.07459/) (2025)
- [Predicting mutational effects on protein binding from folding energy](https://mlsanity.com/paper/2507.05502/) (2025)
- [Zero-shot protein stability prediction by inverse folding models: a free energy interpretation](https://mlsanity.com/paper/2506.05596/) (2025)
- [EnerBridge-DPO: Energy-Guided Protein Inverse Folding with Markov Bridges and Direct Preference Optimization](https://mlsanity.com/paper/2506.09496/) (2025)
- [AlphaFold Database Debiasing for Robust Inverse Folding](https://mlsanity.com/paper/2506.08365/) (2025)

## Abstract (arXiv)

Predicting the change in binding free energy (Delta Delta G) is crucial for understanding and modulating protein-protein interactions, which are critical in drug design. Due to the scarcity of experimental Delta Delta G data, existing methods focus on pre-training, while neglecting the importance of alignment. In this work, we propose the Boltzmann Alignment technique to transfer knowledge from pre-trained inverse folding models to Delta Delta G prediction. We begin by analyzing the thermodynamic definition of Delta Delta G and introducing the Boltzmann distribution to connect energy with protein conformational distribution. However, the protein conformational distribution is intractable; therefore, we employ Bayes' theorem to circumvent direct estimation and instead utilize the log-likelihood provided by protein inverse folding models for Delta Delta G estimation. Compared to previous inverse folding-based methods, our method explicitly accounts for the unbound state of protein complex in the Delta Delta G thermodynamic cycle, introducing a physical inductive bias and achieving both supervised and unsupervised state-of-the-art (SoTA) performance. Experimental results on SKEMPI v2 indicate that our method achieves Spearman coefficients of 0.3201 (unsupervised) and 0.5134 (supervised), significantly surpassing the previously reported SoTA values of 0.2632 and 0.4324, respectively. Futhermore, we demonstrate the capability of our method on binding energy prediction, protein-protein docking and antibody optimization tasks.