Matthew A. Spence

This study evaluates ML architectures for protein fitness prediction using simulated and empirical landscapes, identifying landscape ruggedness as a key determinant of accuracy.

This study uses ancestral sequence reconstruction to explore PET-degrading cutinase sequence space, revealing diverse evolutionary pathways and the effectiveness of LASE in mapping fitness landscapes.

The paper introduces a graph-based measure of protein fitness landscape ruggedness and shows it concentrates at regions where dominant constraints change.

The paper examines how thermodynamic binding strategies in proteins evolved in response to temperature changes, showing a shift from entropy-driven to enthalpy-driven binding over time.

This paper introduces LASE, a method using ancestral sequence reconstruction to train protein language models with higher accuracy and faster computation than large models.

The paper performs a comprehensive phylogenetic analysis of ribonucleotide reductase (RNR) family to identify an ancestral clade and trace evolutionary diversification through insertions and extensions.

This study constructs a comprehensive phylogeny of 6000 serpin sequences across all kingdoms of life, showing convergent evolution rather than horizontal gene transfer for the central hub proteins.