Protein structure prediction is one of the most important problems in computational biology. The most successful computational approach, also called template-based modeling, identifies templates with solved crystal structures for the query proteins and constructs three dimensional models based on sequence/structure alignments. Although substantial effort has been made to improve protein sequence alignment, the accuracy of alignments between distantly related proteins is still unsatisfactory. In this thesis, I will introduce a number of statistical machine learning methods to build accurate alignments between a protein sequence and its template structures, especially for proteins having only distantly related templates. For a protein with only one good template, we develop a regression-tree based Conditional Random Fields (CRF) model for pairwise protein sequence/structure alignment. By learning a nonlinear threading scoring function, we are able to leverage the correlation among different sequence and structural features. We also introduce an information-theoretic measure to guide the learning algorithm to better exploit the structural features for low-homology proteins with little evolutionary information in their sequence profile. For a protein with multiple good templates, we design a probabilistic consistency approach to thread the protein to all templates simultaneously. By minimizing the discordance between the pairwise alignments of the protein and templates, we are able to construct a multiple sequence/structure alignment, which leads to better structure predictions than any single-template based prediction.
This chapter deals with approaches for protein three-dimensional structure prediction, starting out from a single input sequence with unknown struc- ture, the 'query' or 'target' sequence. Both template based and template free modelling techniques are treated, and how resulting structural models may be selected and refined. We give a concrete flowchart for how to de- cide which modelling strategy is best suited in particular circumstances, and which steps need to be taken in each strategy. Notably, the ability to locate a suitable structural template by homology or fold recognition is crucial;…
Structural templates are 3D signatures representing protein functional sites, such as ligand binding cavities, metal coordination motifs or catalytic sites. Here we explore methods to generate template libraries and algorithms to query structures for conserved 3D motifs. Applications of templates are discussed, as well as some exemplar cases for examining evolutionary links in enzymes. We also introduce the concept of using more than one template per structure to represent flexible sites, as an approach to better understand catalysis through snapshots captured in enzyme structures. Functional …
Prediction of protein structures using computational approaches has been explored for over two decades, paving a way for more focused research and development of algorithms in comparative modelling, ab intio modelling and structure refinement protocols. A tremendous success has been witnessed in template-based modelling protocols, whereas strategies that involve template-free modelling still lag behind, specifically for larger proteins (> 150 a.a.). Various improvements have been observed in ab initio protein structure prediction methodologies overtime, with recent ones attributed to the usage…
This chapter gives a graceful introduction to problem of protein three- dimensional structure prediction, and focuses on how to make structural sense out of a single input sequence with unknown structure, the 'query' or 'target' sequence. We give an overview of the different classes of modelling techniques, notably template-based and template free. We also discuss the way in which structural predictions are validated within the global com- munity, and elaborate on the extent to which predicted structures may be trusted and used in practice. Finally we discuss whether the concept of a sin- gle …
Protein structure prediction remains to be an open problem in bioinformatics. There are two main categories of methods for protein structure prediction: Free Modeling (FM) and Template Based Modeling (TBM). Protein threading, belonging to the category of template based modeling, identifies the most likely fold with the target by making a sequence-structure alignment between target protein and template protein. Though protein threading has been shown to more be successful for protein structure prediction, it performs poorly for remote homology detection.
Protein tertiary structure prediction has improved dramatically in recent years. A considerable fraction of various proteomes can be modelled in the absence of structural templates. We ask whether our DMPfold method can model all the proteins without templates in the JCVI-syn3.0 minimal genome, which contains 438 proteins. We find that a useful tertiary structure annotation can be provided for all but 10 proteins. The models may help annotate function in cases where it is unknown, and provide coverage for 29 predicted protein-protein interactions which lacked monomer models. We also show that …
Motivation Protein fold recognition is an important problem in structural bioinformatics. Almost all traditional fold recognition methods use sequence (homology) comparison to indirectly predict the fold of a tar get protein based on the fold of a template protein with known structure, which cannot explain the relationship between sequence and fold. Only a few methods had been developed to classify protein sequences into a small number of folds due to methodological limitations, which are not generally useful in practice. Results We develop a deep 1D-convolution neural network (DeepSF) to dire…
In the template-assistance model, normal prion protein (PrPC), the pathogenic cause of prion diseases such as Creutzfeldt-Jakob (CJD) in human, Bovine Spongiform Encephalopathy (BSE) in cow, and scrapie in sheep, converts to infectious prion (PrPSc) through an autocatalytic process triggered by a transient interaction between PrPC and PrPSc. Conventional studies suggest the S1-H1-S2 region in PrPC to be the template of S1-S2 $β$-sheet in PrPSc, and the conformational conversion of PrPC into PrPSc may involve an unfolding of H1 in PrPC and its refolding into the $β$-sheet in PrPSc. Here we cond…
We discuss a model of protein conformations where the conformations are combinations of short fragments from some small set. For these fragments we consider a distribution of frequencies of occurrence of pairs (sequence of amino acids, conformation), averaged over some balls in the spaces of sequences and conformations. These frequencies can be estimated due to smallness of epsilon-entropy of the set of conformations of protein fragments. We consider statistical potentials for protein fragments which describe the mentioned frequencies of occurrence and discuss model of free energy of a protein…
Proteins are essential for maintaining life. For example, knowing the structure of a protein, cell regulatory mechanisms of organisms can be modeled, supporting the development of disease treatments or the understanding of relationships between protein structures and food attributes. However, discovering the structure of a protein can be a difficult and expensive task, since it is hard to explore the large search to predict even a small protein. Template-based methods (coarse-grained, homology, threading etc) depend on Prior Knowledge (PK) of proteins determined using other methods as X-Ray Cr…
We study statistical properties of interacting protein-like surfaces and predict two strong, related effects: (i) statistically enhanced self-attraction of proteins; (ii) statistically enhanced attraction of proteins with similar structures. The effects originate in the fact that the probability to find a pattern self-match between two identical, even randomly organized interacting protein surfaces is always higher compared with the probability for a pattern match between two different, promiscuous protein surfaces. This theoretical finding explains statistical prevalence of homodimers in prot…
Statistical coupling analysis (SCA) is a method for analyzing multiple sequence alignments that was used to identify groups of coevolving residues termed "sectors". The method applies spectral analysis to a matrix obtained by combining correlation information with sequence conservation. It has been asserted that the protein sectors identified by SCA are functionally significant, with different sectors controlling different biochemical properties of the protein. Here we reconsider the available experimental data and note that it involves almost exclusively proteins with a single sector. We show…
This chapter discusses theoretical framework and methods for developing knowledge-based potential functions essential for protein structure prediction, protein-protein interaction, and protein sequence design. We discuss in some details about the Miyazawa-Jernigan contact statistical potential, distance-dependent statistical potentials, as well as geometric statistical potentials. We also describe a geometric model for developing both linear and non-linear potential functions by optimization. Applications of knowledge-based potential functions in protein-decoy discrimination, in protein-protei…
We report the performance of our newly introduced Ensemble Docking with Enhanced sampling of pocket Shape (EDES) protocol coupled to a template-based algorithm to generate near-native ligand conformations in the 2019 iteration of the Grand Challenge organized by the D3R consortium. Using either AutoDock4.2 or HADDOCK2.2 docking programs (each software in two variants of the protocol) our method generated native-like poses among the top 5 submitted for evaluation for most of the 20 targets with similar performances. The protein selected for GC4 was the human beta-site amyloid precursor protein …
Developing an accurate scoring function is essential for successfully predicting protein structures. In this study, we developed a statistical potential function, called OPUS-Beta, for energetically evaluating beta-sheet contact pattern (the entire residue-residue beta-contacts of a protein) independent of the atomic coordinate information. The OPUS-Beta potential contains five terms, i.e., a self-packing term, a pairwise inter-strand packing term, a pairwise intra-strand packing term, a lattice term and a hydrogen-bonding term. The results show that, in recognizing the native beta-contact pat…
Studying evolutionary correlations in alignments of homologous sequences by means of an inverse Potts model has proven useful to obtain residue-residue contact energies and to predict contacts in proteins. The quality of the results depend much on several choices of the detailed model and on the algorithms used. We built, in a very controlled way, synthetic alignments with statistical properties similar to those of real proteins, and used them to assess the performance of different inversion algorithms and of their variants. Realistic synthetic alignments display typical features of low--tempe…
X-ray crystallography is the predominant method for obtaining atomic-scale information about biological macromolecules. Despite the success of the technique, obtaining well diffracting crystals still critically limits going from protein to structure. In practice, the crystallization process proceeds through knowledge-informed empiricism. Better physico-chemical understanding remains elusive because of the large number of variables involved, hence little guidance is available to systematically identify solution conditions that promote crystallization. To help determine relationships between mac…
A newly developed statistical pair potential based on Distance-scaled Finite Ideal-gas REference (DFIRE) state is applied to unbound protein-protein docking structure selections. The performance of the DFIRE energy function is compared to those of the well-established ZDOCK energy scores and RosettaDock energy function using the comprehensive decoy sets generated by ZDOCK and RosettaDock. Despite significant difference in the functional forms and complexities of the three energy scores, the differences in overall performance for docking structure selections are small between DFIRE and ZDOCK2.3…
The SARS-CoV-2 spike (S) protein facilitates viral infection, and has been the focus of many structure determination efforts. Its flexible loop regions are known to be involved in protein binding and may adopt multiple conformations. This paper identifies the S protein loops and studies their conformational variability based on the available Protein Data Bank (PDB) structures. While most loops had essentially one stable conformation, 17 of 44 loop regions were observed to be structurally variable with multiple substantively distinct conformations based on a cluster analysis. Loop modeling meth…
The authors address the problem of downhill protein folding in the framework of a simple statistical mechanical model, which allows an exact solution for the equilibrium and a semianalytical treatment of the kinetics. Focusing on protein 1BBL, a candidate for downhill folding behavior, and comparing it to the WW domain of protein PIN1, a two-state folder of comparable size, the authors show that there are qualitative differences in both the equilibrium and kinetic properties of the two molecules. However, the barrierless scenario which would be expected if 1BBL were a true downhill folder is o…
A pharmacophore consists of the parts of the structure of the ligand that are sufficient to express the biological and pharmacological effects of the ligand. It is usually a substructure of the entire structure of the ligand. Small organic molecules called ligands or metabolites in the cell form complexes with biomolecules (usually proteins) to serve different purposes. The sites at which the ligands bind are known as ligand binding sites, which are essentially "pockets" which have complementary shapes and patterns of charge distribution with the ligands. Sometimes a pocket is induced by the l…
As the major target of many vaccines and neutralizing antibodies against SARS-CoV-2, the spike (S) protein is observed to mutate over time. In this paper, we present statistical approaches to tackle some challenges associated with the analysis of S-protein data. We build a Bayesian hierarchical model to study the temporal and spatial evolution of S-protein sequences, after grouping the sequences into representative clusters. We then apply sampling methods to investigate possible changes to the S-protein's 3-D structure as a result of commonly observed mutations. While the increasing spread of …
We present a novel statistical mechanics formalism for the theoretical description of the process of protein folding$\leftrightarrow$unfolding transition in water environment. The formalism is based on the construction of the partition function of a protein obeying two-stage-like folding kinetics. Using the statistical mechanics model of solvation of hydrophobic hydrocarbons we obtain the partition function of infinitely diluted solution of proteins in water environment. The calculated dependencies of the protein heat capacities upon temperature are compared with the corresponding results of e…
The statistical properties of protein folding within the ϕ^4 model are investigated. The calculation is performed using statistical mechanics and path integral method. In particular, the evolution of heat capacity in term of temperature is given for various levels of the nonlinearity of source and the strength of interaction between protein backbone and nonlinear source. It is found that the nonlinear source contributes constructively to the specific heat especially at higher temperature when it is weakly interacting with the protein backbone. This indicates increasing energy absorption as the…
George Mason University（乔治梅森大学）；Los Alamos National Laboratory（洛斯阿拉莫斯国家实验室）
Identifying one or more biologically-active/native decoys from millions of non-native decoys is one of the major challenges in computational structural biology. The extreme lack of balance in positive and negative samples (native and non-native decoys) in a decoy set makes the problem even more complicated. Consensus methods show varied success in handling the challenge of decoy selection despite some issues associated with clustering large decoy sets and decoy sets that do not show much structural similarity. Recent investigations into energy landscape-based decoy selection approaches show pr…
The analysis of correlations of amino acid occurrences in globular proteins has led to the development of statistical tools that can identify native contacts -- portions of the chains that come to close distance in folded structural ensembles. Here we introduce a statistical coupling analysis for repeat proteins -- natural systems for which the identification of domains remains challenging. We show that the inherent translational symmetry of repeat protein sequences introduces a strong bias in the pair correlations at precisely the length scale of the repeat-unit. Equalizing for this bias reve…
Modern biomedicine is challenged to predict the effects of genetic variation. Systematic functional assays of point mutants of proteins have provided valuable empirical information, but vast regions of sequence space remain unexplored. Fortunately, the mutation-selection process of natural evolution has recorded rich information in the diversity of natural protein sequences. Here, building on probabilistic models for correlated amino-acid substitutions that have been successfully applied to determine the three-dimensional structures of proteins, we present a statistical approach for quantifyin…
Local protein structure analysis is informative to protein structure analysis and has been used successfully in protein structure prediction and others. Proteins have recurring structural features, such as helix caps and beta turns, which often have strong amino acid sequence preferences. And the challenges for local structure analysis have been identification and assignment of such common short structural motifs. This paper proposes a new mathematical framework that can be applied to analysis of the local structure of proteins, where local conformations of protein backbones are described usin…
Integral membrane proteins deform the surrounding bilayer creating long-ranged forces that influence distant proteins. These forces can be attractive or repulsive, depending on the proteins' shape, height, contact angle with the bilayer, as well as the local membrane curvature. Although interaction energies are not pairwise additive, for sufficiently low protein density, thermodynamic properties depend only upon pair interactions. Here, we compute pair interaction potentials and entropic contributions to the two-dimensional osmotic pressure of a collection of noncircular proteins. In contrast …
Statistical-mechanical lattice models for protein-DNA binding are well established as a method to describe complex ligand binding equilibriums measured in vitro with purified DNA and protein components. Recently, a new field of applications has opened up for this approach since it has become possible to experimentally quantify genome-wide protein occupancies in relation to the DNA sequence. In particular, the organization of the eukaryotic genome by histone proteins into a nucleoprotein complex termed chromatin has been recognized as a key parameter that controls the access of transcription fa…
Within bioinformatics, the textual alignment of amino acid sequences has long dominated the determination of similarity between proteins, with all that implies for shared structure, function and evolutionary descent. Despite the relative success of modern-day sequence alignment algorithms, so-called alignment-free approaches offer a complementary means of determining and expressing similarity, with potential benefits in certain key applications, such as regression analysis of protein structure-function studies, where alignment-base similarity has performed poorly. Here, we offer a fresh, stati…
Toyota Technological Institute at Chicago（芝加哥丰田技术研究所）
Recently exciting progress has been made on protein contact prediction, but the predicted contacts for proteins without many sequence homologs is still of low quality and not very useful for de novo structure prediction. This paper presents a new deep learning method that predicts contacts by integrating both evolutionary coupling (EC) and sequence conservation information through an ultra-deep neural network formed by two deep residual networks. This deep neural network allows us to model very complex sequence-contact relationship as well as long-range inter-contact correlation. Our method gr…
A grand canonical Monte Carlo (MC) algorithm is presented for studying the lattice gas model (LGM) of multiple protein sequence alignment, which coherently combines long-range interactions and variable-length insertions. MC simulations are used for both parameter optimization of the model and production runs to explore the sequence subspace around a given protein family. In this Note, I describe the details of the MC algorithm as well as some preliminary results of MC simulations with various temperatures and chemical potentials, and compare them with the mean-field approximation. The existenc…
In the course of evolution, proteins undergo important changes in their amino acid sequences, while their three-dimensional folded structure and their biological function remain remarkably conserved. Thanks to modern sequencing techniques, sequence data accumulate at unprecedented pace. This provides large sets of so-called homologous, i.e.~evolutionarily related protein sequences, to which methods of inverse statistical physics can be applied. Using sequence data as the basis for the inference of Boltzmann distributions from samples of microscopic configurations or observables, it is possible…
Protein aggregation is an important field of investigation because it is closely related to the problem of neurodegenerative diseases, to the development of biomaterials, and to the growth of cellular structures such as cyto-skeleton. Self-aggregation of protein amyloids, for example, is a complicated process involving many species and levels of structures. This complexity, however, can be dealt with using statistical mechanical tools, such as free energies, partition functions, and transfer matrices. In this article, we review general strategies for studying protein aggregation using statisti…
Proteins are biomolecules of life. They fold into a great variety of three-dimensional (3D) shapes. Underlying these folding patterns are many recurrent structural fragments or building blocks (analogous to `LEGO bricks'). This paper reports an innovative statistical inference approach to discover a comprehensive dictionary of protein structural building blocks from a large corpus of experimentally determined protein structures. Our approach is built on the Bayesian and information-theoretic criterion of minimum message length. To the best of our knowledge, this work is the first systematic an…
Proteins are polymers of amino acids. These macromolecules are synthesized by intracellular machines called ribosomes. Although the experimental investigation of protein synthesis has been a traditional area of research in molecular cell biology, important quantitative models of protein synthesis have been reported in research journals devoted to statistical physics and related interdisciplinary topics. From the perspective of a physicist, protein synthesis is the classical transport of interacting ribosomes on a messenger RNA (mRNA) template that dictates the sequence of the amino acids on th…
Identifying protein-protein interactions is crucial for a systems-level understanding of the cell. Recently, algorithms based on inverse statistical physics, e.g. Direct Coupling Analysis (DCA), have allowed to use evolutionarily related sequences to address two conceptually related inference tasks: finding pairs of interacting proteins, and identifying pairs of residues which form contacts between interacting proteins. Here we address two underlying questions: How are the performances of both inference tasks related? How does performance depend on dataset size and the quality? To this end, we…
In a statistical approach to protein structure analysis, Miyazawa and Jernigan (MJ) derived a $20\times 20$ matrix of inter-residue contact energies between different types of amino acids. Using the method of eigenvalue decomposition, we find that the MJ matrix can be accurately reconstructed from its first two principal component vectors as $M_{ij}=C_0+C_1(q_i+q_j)+C_2 q_i q_j$, with constant $C$'s, and 20 $q$ values associated with the 20 amino acids. This regularity is due to hydrophobic interactions and a force of demixing, the latter obeying Hildebrand's solubility theory of simple liquid…
$α$-helices stand out as common and relatively invariant secondary structural elements of proteins. However, $α$-helices are not rigid bodies and their deformations can be significant in protein function ({\it e.g.} coiled coils). To quantify the flexibility of $α$-helices we have performed a structural principal-component analysis of helices of different lengths from a representative set of protein folds in the Protein Data Bank. We find three dominant modes of flexibility: two degenerate bend modes and one twist mode. The data are consistent with independent Gaussian distributions for each m…
Statistical analysis of evolutionary-related protein sequences provides insights about their structure, function, and history. We show that Restricted Boltzmann Machines (RBM), designed to learn complex high-dimensional data and their statistical features, can efficiently model protein families from sequence information. We here apply RBM to twenty protein families, and present detailed results for two short protein domains, Kunitz and WW, one long chaperone protein, Hsp70, and synthetic lattice proteins for benchmarking. The features inferred by the RBM are biologically interpretable: they ar…
We study the thermodynamical properties of a topology-based model proposed by Galzitskaya and Finkelstein for the description of protein folding. We devise and test three different mean-field approaches for the model, that simplify the treatment without spoiling the description. The validity of the model and its mean-field approximations is checked by applying them to the $β$-hairpin fragment of the immunoglobulin-binding protein (GB1) and making a comparison with available experimental data and simulation results. Our results indicate that this model is a rather simple and reasonably good too…
PHAST is a software package written in standard Fortran, with MPI and CUDA extensions, able to efficiently perform parallel multicanonical Monte Carlo simulations of single or multiple heteropolymeric chains, as coarse-grained models for proteins. The outcome data can be straightforwardly analyzed within its microcanonical Statistical Thermodynamics module, which allows for computing the entropy, caloric curve, specific heat and free energies. As a case study, we investigate the aggregation of heteropolymers bioinspired on $Aβ_{25-33}$ fragments and their cross-seeding with $IAPP_{20-29}$ isof…
The multiple sequence alignment (MSA) of a protein family provides a wealth of information in terms of the conservation pattern of amino acid residues not only at each alignment site but also between distant sites. In order to statistically model the MSA incorporating both short-range and long-range correlations as well as insertions, I have derived a lattice gas model of the MSA based on the principle of maximum entropy. The partition function, obtained by the transfer matrix method with a mean-field approximation, accounts for all possible alignments with all possible sequences. The model pa…
We present a new method to extract distance and orientation dependent potentials between amino acid side chains using a database of protein structures and the standard Boltzmann device. The importance of orientation dependent interactions is first established by computing orientational order parameters for proteins with alpha-helical and beta-sheet architecture. Extraction of the anisotropic interactions requires defining local reference frames for each amino acid that uniquely determine the coordinates of the neighboring residues. Using the local reference frames and histograms of the radial …
In this paper, we introduce an approach to the protein folding problem from the point of view of statistical physics. Protein folding is a stochastic process by which a polypeptide folds into its characteristic and functional 3D structure from random coil. The process involves an intricate interplay between global geometry and local structure, and each protein seems to present special problems. We introduce CSAW (conditioned self-avoiding walk), a model of protein folding that combines the features of self-avoiding walk (SAW) and the Monte Carlo method. In this model, the unfolded protein chai…
During the synthesis of integral membrane proteins (IMPs), the hydrophobic amino acids of the polypeptide sequence are partitioned mostly into the membrane interior and hydrophilic amino acids mostly into the aqueous exterior. We analyze the minimum free energy state of polypeptide sequences partitioned into alpha-helical transmembrane (TM) segments and the role of thermal fluctuations using a many-body statistical mechanics model. Results suggest that IMP TM segment partitioning shares important features with general theories of protein folding. For random polypeptide sequences, the minimum f…
Inverse statistical approaches to determine protein structure and function from Multiple Sequence Alignments (MSA) are emerging as powerful tools in computational biology. However the underlying assumptions of the relationship between the inferred effective Potts Hamiltonian and real protein structure and energetics remain untested so far. Here we use lattice protein model (LP) to benchmark those inverse statistical approaches. We build MSA of highly stable sequences in target LP structures, and infer the effective pairwise Potts Hamiltonians from those MSA. We find that inferred Potts Hamilto…
We consider a general incompressible finite model protein of size M in its environment, which we represent by a semiflexible copolymer consisting of amino acid residues classified into only two species (H and P, see text) following Lau and Dill. We allow various interactions between chemically unbonded residues in a given sequence and the solvent (water), and exactly enumerate the number of conformations W(E) as a function of the energy E on an infinite lattice under two different conditions: (i) we allow conformations that are restricted to be compact (known as Hamilton walk conformations), a…
Functional protein-protein interactions are crucial in most cellular processes. They enable multi-protein complexes to assemble and to remain stable, and they allow signal transduction in various pathways. Functional interactions between proteins result in coevolution between the interacting partners, and thus in correlations between their sequences. Pairwise maximum-entropy based models have enabled successful inference of pairs of amino-acid residues that are in contact in the three-dimensional structure of multi-protein complexes, starting from the correlations in the sequence data of known…
This work demonstrates how a complete statistical model quantifying the evolution of pairs of aligned proteins can be constructed from a time-parameterised substitution matrix and a time-parameterised 3-state alignment machine. All parameters of such a model can be inferred from any benchmark data-set of aligned protein sequences. This allows us to examine nine well-known substitution matrices on six benchmarks curated using various structural alignment methods; any matrix that does not explicitly model a "time"-dependent Markov process is converted to a corresponding base-matrix that does. In…
We study the statistical properties of hydrophobic/polar model sequences with unique native states on the square lattice. It is shown that this ensemble of sequences differs from random sequences in significant ways in terms of both the distribution of hydrophobicity along the chains and total hydrophobicity. Whenever statistically feasible, the analogous calculations are performed for a set of real enzymes, too.
From the analysis of sizes of approximately 130 small icosahedral viruses we find that there is a typical structural capsid protein, having a mean diameter of 5 nm and a mean thickness of 3 nm, with more than two thirds of the analyzed capsid proteins having thicknesses between 2 nm and 4 nm. To investigate whether, in addition to the fairly conserved geometry, capsid proteins show similarities in the way they interact with one another, we examined the shapes of the capsids in detail. We classified them numerically according to their similarity to sphere and icosahedron and an interpolating se…
