EHCube4P: Learning Epistatic Patterns Through Hypercube Graph Convolution Neural Network for Protein Fitness Function Estimation

TL;DR

This paper introduces EHCube4P, a novel graph convolutional neural network framework that models protein sequence landscapes as hypercubes, integrating wavelet denoising to improve fitness prediction accuracy in noisy, complex landscapes.

Contribution

The study presents a new hypercube-based GCN model combined with wavelet denoising for robust protein fitness prediction, capturing higher-order epistatic interactions beyond pairwise effects.

Findings

Effective noise suppression with wavelet transform.

Generalizes well across different enzyme datasets.

Higher accuracy on smoother fitness landscapes.

Abstract

Understanding the relationship between protein sequences and their functions is fundamental to protein engineering, but this task is hindered by the combinatorially vast sequence space and the experimental noise inherent in fitness measurements. In this study, we present a novel framework that models the sequence landscape as a hypercube $H(k,2)$ and integrates wavelet-based signal denoising with a graph convolutional neural network (GCN) to predict protein fitness across rugged fitness landscapes. Using a dataset of 419 experimentally measured mutant sequences of the Tobacco 5-Epi-Aristolochene Synthase (TEAS) enzyme, we preprocess the fitness signals using a 1-D discrete wavelet transform with a Daubechies-3 basis to suppress experimental noise while preserving local epistatic patterns. Our model comprises two GCN layers, allowing for beyond pairwise aggregation, followed by a multi-layer perceptron (MLP). We show that our approach, EHCube4P, generalizes well across different enzyme activity datasets and effectively captures higher-order mutational interactions. Performance varies with the ruggedness of the fitness landscape, with smoother signals yielding higher test set $r^2$ scores. These results demonstrate that combining wavelet preprocessing with graph-based deep learning enhances the robustness and generalization of fitness prediction, particularly for sparse and noisy biological datasets. The approach provides a scalable and interpretable framework for protein fitness estimation applicable to a broad range of combinatorial biological systems.