Profile Conditional Random Fields for Modeling Protein Families with Structural Information

TL;DR

This paper introduces profile conditional random fields (CRFs), a novel statistical model for protein families that integrates sequence correlations and long-range interactions, improving upon traditional profile HMMs.

Contribution

The paper presents a new profile CRF model that incorporates arbitrary correlations and long-range interactions, extending existing models like profile HMMs and FR theory.

Findings

Model formulation with correlations and long-range interactions

Algorithms for partition functions and marginal probabilities

Bayesian framework for parameter learning

Abstract

A statistical model of protein families, called profile conditional random fields (CRFs), is proposed. This model may be regarded as an integration of the profile hidden Markov model (HMM) and the Finkelstein-Reva (FR) theory of protein folding. While the model structure of the profile CRF is almost identical to the profile HMM, it can incorporate arbitrary correlations in the sequences to be aligned to the model. In addition, like in the FR theory, the profile CRF can incorporate long-range pairwise interactions between model states via mean-field-like approximations. We give the detailed formulation of the model, self-consistent approximations for treating long-range interactions, and algorithms for computing partition functions and marginal probabilities. We also outline the methods for the global optimization of model parameters as well as a Bayesian framework for parameter learning and selection of optimal alignments.