Protein 3D Structure Computed from Evolutionary Sequence Variation

Abstract

The evolutionary trajectory of a protein through sequence space is constrained by its function. Collections of sequence homologs record the outcomes of millions of evolutionary experiments in which the protein evolves according to these constraints. Deciphering the evolutionary record held in these sequences and exploiting it for predictive and engineering purposes presents a formidable challenge. The potential benefit of solving this challenge is amplified by the advent of inexpensive high-throughput genomic sequencing.In this paper we ask whether we can infer evolutionary constraints from a set of sequence homologs of a protein. The challenge is to distinguish true co-evolution couplings from the noisy set of observed correlations. We address this challenge using a maximum entropy model of the protein sequence, constrained by the statistics of the multiple sequence alignment, to infer residue pair couplings. Surprisingly, we find that the strength of these inferred couplings is an excellent predictor of residue-residue proximity in folded structures. Indeed, the top-scoring residue couplings are sufficiently accurate and well-distributed to define the 3D protein fold with remarkable accuracy.We quantify this observation by computing, from sequence alone, all-atom 3D structures of fifteen test proteins from different fold classes, ranging in size from 50 to 260 residues., including a G-protein coupled receptor. These blinded inferences are de novo, i.e., they do not use homology modeling or sequence-similar fragments from known structures. The co-evolution signals provide sufficient information to determine accurate 3D protein structure to 2.7–4.8 Å Cα-RMSD error relative to the observed structure, over at least two-thirds of the protein (method called EVfold, details at http://EVfold.org). This discovery provides insight into essential interactions constraining protein evolution and will facilitate a comprehensive survey of the universe of protein structures, new strategies in protein and drug design, and the identification of functional genetic variants in normal and disease genomes.

Structure of the peptidoglycan polymerase RodA resolved by evolutionary coupling analysis

“…A total of 220 folded models for T. thermophilus RodA were generated for increasing numbers of EC restraints with using the folding protocol in EVfold 9 which itself uses a distance geometry and simulated annealing protocol in CNS 32,33 . All models were ranked as described in previously 11,34 and the 50 top-ranked models from each of the MSAs were used as as molecular replacement search models ( vide infra ). The full EVfold software package and ReadMe is available at https://github.com/debbiemarkslab/EVcouplings.…”

“…Recent methodological advances in molecular replacement have expanded the range of suitable templates 8 , and evolutionary co-variation analysis now allows for fold prediction even in the absence of prior structural data 9–11 . This approach exploits the fact that residues that interact with one another structurally tend to co-evolve to maintain their interactions.…”

HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment

“…HHblits has the potential to improve many downstream analysis and prediction methods, such as a de novo protein structure prediction method requiring large and accurate MSAs 16 . methods Methods and any associated references are available in the online version of the paper at http://www.nature.com/naturemethods/.…”

Large pseudocounts andL2-norm penalties are necessary for the mean-field inference of Ising and Potts models

“…The rms error captures the absolute difference between the true and inferred couplings but is unable to clearly distinguish whether the relative ordering of the couplings has been correctly inferred. This limitation is problematic since many practical applications, such as the prediction of protein contacts from MF inference on sequence data [4,5,34], rely on proper rank ordering of the inferred couplings rather than their absolute magnitude. Information about the correct rank ordering can be determined from the rank correlation between the true and inferred couplings.…”