Q-BIOLAT: Binary Latent Protein Fitness Landscapes for QUBO-Based Optimization

arXiv cs.LG / 3/31/2026

📰 NewsSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

Q-BIOLAT is a framework that models protein fitness landscapes using compact binary latent spaces and learns a QUBO surrogate capturing unary and pairwise interactions from pretrained protein language model embeddings.
The work highlights that representation quality matters for optimization: autoencoder-based latent spaces can collapse after binarization, creating degenerate landscapes that break combinatorial search, while PCA-like structured representations remain high-entropy and decodable.
Experiments across multiple datasets and data regimes show that classical combinatorial optimizers (simulated annealing, genetic algorithms, and greedy hill climbing) perform strongly in structured binary latent spaces.
By casting the optimization objective into QUBO form, Q-BIOLAT links machine learning representation learning with discrete and quantum-inspired optimization workflows.
The authors provide public code and data via the linked GitHub repository to support reproduction and further research.

Abstract

Protein fitness optimization is inherently a discrete combinatorial problem, yet most learning-based approaches rely on continuous representations and are primarily evaluated through predictive accuracy. We introduce Q-BIOLAT, a framework for modeling and optimizing protein fitness landscapes in compact binary latent spaces. Starting from pretrained protein language model embeddings, we construct binary latent representations and learn a quadratic unconstrained binary optimization (QUBO) surrogate that captures unary and pairwise interactions. Beyond its formulation, Q-BIOLAT provides a representation-centric perspective on protein fitness modeling. We show that representations with similar predictive performance can induce fundamentally different optimization landscapes. In particular, learned autoencoder-based representations collapse after binarization, producing degenerate latent spaces that fail to support combinatorial search, whereas simple structured representations such as PCA yield high-entropy, decodable, and optimization-friendly latent spaces. Across multiple datasets and data regimes, we demonstrate that classical combinatorial optimization methods, including simulated annealing, genetic algorithms, and greedy hill climbing, are highly effective in structured binary latent spaces. By expressing the objective in QUBO form, our approach connects modern machine learning with discrete and quantum-inspired optimization. Our implementation and dataset are publicly available at: https://github.com/HySonLab/Q-BIOLAT-Extended