Hash Collisions in Molecular Fingerprints: Effects on Property Prediction and Bayesian Optimization

TL;DR

This paper examines how hash collisions in molecular fingerprints affect property prediction and Bayesian optimization, finding that exact fingerprints slightly improve prediction accuracy but do not significantly enhance optimization results.

Contribution

It provides an empirical comparison between exact and compressed fingerprints, highlighting their impact on molecular property prediction and Bayesian optimization.

Findings

Exact fingerprints improve prediction accuracy slightly.

Hash collisions have limited impact on Bayesian optimization.

Standard compressed fingerprints are generally sufficient for optimization tasks.

Abstract

Molecular fingerprinting methods use hash functions to create fixed-length vector representations of molecules. However, hash collisions cause distinct substructures to be represented with the same feature, leading to overestimates in molecular similarity calculations. We investigate whether using exact fingerprints improves accuracy compared to standard compressed fingerprints in molecular property prediction and Bayesian optimization where the underlying predictive model is a Gaussian process. We find that using exact fingerprints yields a small yet consistent improvement in predictive accuracy on five molecular property prediction benchmarks from the DOCKSTRING dataset. However, these gains did not translate to significant improvements in Bayesian optimization performance.