Exploring protein-protein interactions using high-throughput datasets and deep learning

Abstract

Protein-protein interactions (PPIs) are fundamental to cellular function.Understanding which proteins interact—and how sequence variation alters these interactions—is essential for advancing therapeutic discovery and protein engineering. High-throughput sequencing technologies enable the large-scale measurement of PPIs, but the resulting datasets are complex and require error correction, modeling, and interpretation to yield meaningful insights. This thesis presents work across the process of designing, executing, and making use of high-throughput data, including (1) designing and modeling mutant protein libraries for large-scale PPI measurement, (2) developing PPI-specific sequencing analysis pipelines, (3) training models on limited structural features for PPI prediction for specific families, and (3) applying feature attribution techniques to interpret sequence-to-function models. Together, this work supports the continued development of experimental and computational tools to deepen our understanding of protein-protein interactions.