Proteins play a central role in biological processes. Yet, sequence-based methods often fall short in capturing the evolutionary nuances critical for understanding their function and the consequences of genetic variations. The challenge arises from the limited perspective provided by conventional multiple sequence alignments, which overlook the rich contextual information inherent in evolutionary trees and ancestral relationships. To address this, we developed a comprehensive framework that first harnesses the power of phylogenetic analysis and ancestral reconstruction to assess the functional status of amino acids. By integrating these evolutionary insights with advanced machine learning techniques, our approach enhances mutation pathogenicity prediction by capturing complex, non-linear relationships. Moreover, a similar tree-based approach refines the detection of authentic co-evolutionary signals, thereby deepening our understanding of protein structure and dynamics. Ultimately, this evolution-informed strategy is applied to elucidate the functional roles of key residues in proteins, linking conservation patterns directly to activation mechanisms and the molecular foundation of disorders.
Events occurring close in time are often linked in memory, providing a framework for those memories. Using in vivo longitudinal imaging of neuronal somas, dendrites, and spines along with activity-dependent manipulations of these compartments, we show that organization of multiple memories is not only dependent on neuronal overlap in the retrosplenial cortex, but also on branch-specific dendritic allocation mechanisms. These results reveal a novel set of rules that govern how linked, and independent memories are allocated to dendritic compartments.
Last update: 2/26/2025, Ralf Bundschuh