Much of the material in “Statistical Mechanics of Chain Molecules” is deeply embedded in my long-term memory. There are two aspects of the work in that book that have been important for my own research. First, is the concept of empirical energies that Flory used to differentiate among different polymer types, In his book energies are given in the form of statistical weights for the different local conformations, and these weights were derived by in fitting experimental information. Much of my career was spent studying proteins and nucleic acids in terms of their empirical interaction energies. The experimental data that we have used to defin “statistical weights” are non-local and are based on the frequencies of occurrences of interacting amino acid types. Globular proteins and RNA are densely packed states with strong long-range interactions, so we have derived many-body potentials and shown how to derivethese if there are sufficient numbers of structures. Recently we derived entropies for the interactions showing that the same charge pairs have the highest entropies and the hydrophobic pairs the lowest entropies, yielding a strong entropy-energy compensation. The second topic is investigating the correlations of sequence positions in multiple sequence alignments, for the ~200 million available protein sequences. These resemble the bond correlations in the Flory book, and the ways that they were formulated to represent the distance distributions in polymers; for proteins the sequence correlations yield significant information about how the structures and sequences are related.

Empirical energies and intramolecular bond and sequence correlations

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.