Augmenting Biological Fitness Prediction Benchmarks with Landscapes Features from GraphFLA

Abstract

Machine learning models increasingly map biological sequence-fitness landscapes to predict mutational effects. Effective performance evaluation of these models demands comprehensive benchmarks curated from empirical data. Despite their impressive scale, existing benchmarks lack topographical information regarding the underlying fitness landscapes, which hampers interpretation and comparison of model performance beyond simple averaged scores. To address this, here we present GraphFLA, a Python framework that constructs and analyzes fitness landscapes from mutagenesis data in diverse sequence modalities (e.g., DNA, RNA, protein and beyond) with up to millions of mutants. GraphFLA calculates a holistic set of 20 biologically relevant features that characterize 4 fundamental aspects of landscape topography: ruggedness, epistasis, navigability and neutrality. By applying GraphFLA to over 5,300 empirical landscapes from ProteinGym, RNAGym, and CIS-BP, we demonstrate its utility in interpreting and comparing the performance of dozens of fitness prediction models, highlighting factors influencing model accuracy and respective advantages of different models. All the resources are available at https://github.com/COLA-Laboratory/GraphFLA.