Structural Principles of Protein Evolvability

This thesis examines the relationship between conditional neutrality, protein sequence encoding, and the role of evolutionary history in shaping protein adaptability. By analyzing algorithmically designed synthetic variants of the ligand-binding protein PSD95pdz3 – where only the core epistatic unit was preserved, and the surrounding constraints were scrambled – I show that these surrounding constraints are essential for adaptation to new functional challenges. This finding, that certain sequence constraints are crucial for evolution, combined with the understanding that these constraints are themselves shaped by evolutionary history, suggests that a protein's past influences its capacity for adaptation. Preliminary results using a continuous evolution system (BTH-PACE) support this, demonstrating that different rates of environmental fluctuation lead to distinct patterns of constraints and varying abilities to generate conditional neutrality for alternate ligands. This underscores the role that evolutionary history has in shaping the pattern of sequence constraints on proteins. To further investigate evolutionary dynamics, I introduce HiDenSeq, a novel method for quantifying key evolutionary parameters using the Luria-Delbrück distribution. These parameters include the distribution of fitness effects, selection pressure, and ability to generate conditional neutrality. Together, these findings offer new insights and tools for understanding the constraints that govern protein evolution.