arXiv subjects

Ronald M. Levy

Publications and source records attributed to Ronald M. Levy.

5 recordsLinked to original sources

arXiv2025

Phylogenetic Corrections and Higher-Order Sequence Statistics in Protein Families: The Potts Model vs MSA Transformer

Recent generative learning models applied to protein multiple sequence alignment (MSA) datasets include simple and interpretable physics-based Potts covariation models and other machine learning models such as MSA-Transformer (MSA-T). The best models accurately reproduce MSA statistics induced by the biophysical constraints within proteins, raising the question of which functional forms best model the underlying physics. The Potts model is usually specified by an effective potential including pairwise residue-residue interaction terms, but it has been suggested that MSA-T can capture the effects induced by effective potentials which include more than pairwise interactions and implicitly account for phylogenetic structure in the MSA. Here we compare the ability of the Potts model and MSA-T to reconstruct higher-order sequence statistics reflecting complex biological sequence constraints. We find that the model performance depends greatly on the treatment of phylogenetic relationships between the sequences, which can induce non-biophysical mutational covariation in MSAs. When using explicit corrections for phylogenetic dependencies, we find the Potts model outperforms MSA-T in detecting epistatic interactions of biophysical origin.

physics.bio-ph↗

arXiv2021

Generative Capacity of Probabilistic Protein Sequence Models

Potts models and variational autoencoders (VAEs) have recently gained popularity as generative protein sequence models (GPSMs) to explore fitness landscapes and predict the effect of mutations. Despite encouraging results, quantitative characterization and comparison of GPSM-generated probability distributions is still lacking. It is currently unclear whether GPSMs can faithfully reproduce the complex multi-residue mutation patterns observed in natural sequences arising due to epistasis. We develop a set of sequence statistics to assess the "generative capacity" of three GPSMs of recent interest: the pairwise Potts Hamiltonian, the VAE, and the site-independent model, using natural and synthetic datasets. We show that the generative capacity of the Potts Hamiltonian model is the largest, in that the higher order mutational statistics generated by the model agree with those observed for natural sequences. In contrast, we show that the VAE's generative capacity lies between the pairwise Potts and site-independent models. Importantly, our work measures GPSM generative capacity in terms of higher-order sequence covariation statistics which we have developed, and provides a new framework for evaluating and interpreting GPSM accuracy that emphasizes the role of epistasis.

cs.LG↗

arXiv2019

Influence of Multiple Sequence Alignment Depth on Potts Statistical Models of Protein Covariation

Potts statistical models have become a popular and promising way to analyze mutational covariation in protein Multiple Sequence Alignments (MSAs) in order to understand protein structure, function and fitness. But the statistical limitations of these models, which can have millions of parameters and are fit to MSAs of only thousands or hundreds of effective sequences using a procedure known as inverse Ising inference, are incompletely understood. In this work we predict how model quality degrades as a function of the number of sequences $N$, sequence length $L$, amino-acid alphabet size $q$, and the degree of conservation of the MSA, in different applications of the Potts models: In "fitness" predictions of individual protein sequences, in predictions of the effects of single-point mutations, in "double mutant cycle" predictions of epistasis, and in 3-d contact prediction in protein structure. We show how as MSA depth $N$ decreases an "overfitting" effect occurs such that sequences in the training MSA have overestimated fitness, and we predict the magnitude of this effect and discuss how regularization can help correct for it, use a regularization procedure motivated by statistical analysis of the effects of finite sampling. We find that as $N$ decreases the quality of point-mutation effect predictions degrade least, fitness and epistasis predictions degrade more rapidly, and contact predictions are most affected. However, overfitting becomes negligible for MSA depths of more than a few thousand effective sequences, as often used in practice, and regularization becomes less necessary. We discuss the implications of these results for users of Potts covariation analysis.

q-bio.QM↗

arXiv2009

The AGBNP2 Implicit Solvation Model

The AGBNP2 implicit solvent model, an evolution of the Analytical Generalized Born plus Non-Polar (AGBNP) model we have previously reported, is presented with the aim of modeling hydration effects beyond those described by conventional continuum dielectric representations. A new empirical hydration free energy component based on a procedure to locate and score hydration sites on the solute surface is introduced to model first solvation shell effects, such as hydrogen bonding, which are poorly described by continuum dielectric models. This new component is added to the Generalized Born and non-polar AGBNP models which have been improved with respect to the description of the solute volume description. We have introduced an analytical Solvent Excluding Volume (SEV) model which reduces the effect of spurious high-dielectric interstitial spaces present in conventional van der Waals representations of the solute volume. The AGBNP2 model is parametrized and tested with respect to experimental hydration free energies of small molecules and the results of explicit solvent simulations. Modeling the granularity of water is one of the main design principles employed for the the first shell solvation function and the SEV model, by requiring that water locations have a minimum available volume based on the size of a water molecule. We show that the new volumetric model produces Born radii and surface areas in good agreement with accurate numerical evaluations. The results of Molecular Dynamics simulations of a series of mini-proteins show that the new model produces conformational ensembles in substantially better agreement with reference explicit solvent ensembles than the original AGBNP model with respect to both structural and energetics measures.

q-bio.BM↗

arXiv1995

On finite-size effects in computer simulations using the Ewald potential

We discuss the origin and relevance for computer simulations of a strong finite-size effect that appears when using the Ewald summation formula. It can be understood as arising from a volume-dependent shift of the potential in a finite, periodic box relative to the infinite volume limit. This shift is due to the fact that the ``zero of energy'' for a periodic system cannot be defined by letting the interacting particles be separated by an infinite distance; the correct definition corresponds to setting its $\bbox k=\bbox 0$ Fourier mode to zero. The implications of this effect for computer simulations are discussed.

chem-ph↗