Better Protein Function Prediction by Modeling Survivorship Bias

TL;DR

This paper introduces Evo-PU, a new positive-unlabeled learning framework that models survivorship bias in protein sequence data to improve functional prediction accuracy.

Contribution

Evo-PU incorporates evolutionary mutation knowledge into PU learning, addressing survivorship bias in well-surveilled single-organism sequence data.

Findings

Evo-PU outperforms standard PU, OCC, and protein language models on influenza, RSV, and SARS-CoV-2 prediction tasks.

Evo-PU demonstrates improved accuracy in predicting functional protein variants.

Potential for generalizing the approach to multi-organism datasets with diverse surveillance coverage.

Abstract

Protein sequence data from nature exhibits survivorship bias: we only observe data from those organisms that survive and reproduce, while non-functional protein mutations are eliminated by natural selection. Thus, predicting whether a protein sequence is functional often requires learning from positive examples alone. While positive-unlabeled (PU) learning frameworks offer a generic solution to this problem, existing PU methods ignore the evolutionary processes that shape sequence observability and cause survivorship bias. Consider a sequence that is one mutation away from a commonly-observed protein variant in a well-surveilled organism. If the sequence were functional, it would likely be observed. If it is not observed, this suggests non-functionality. In contrast, sequences that are unlikely to arise through mutation may be missing simply because they never arose. Thus, these two kinds of missing sequences should be treated differently when training models. In this work, we propose Evo-PU, a PU learning framework that uses a scientific understanding of nucleotide mutation to model survivorship bias for well-surveilled single-organism sequence data. On three prediction tasks using single-organism uniform-coverage surveillance data -- predicting results from held-out influenza and respiratory syncytial virus (RSV) mutagenesis studies, and predicting future SARS-CoV-2 variants -- Evo-PU outperforms standard PU learning, one-class classification (OCC), and protein language models (PLMs). On prediction tasks from multi-organism ProteinGym datasets with more heterogeneous surveillance coverage, we identify opportunities to generalize our approach.