DOCSLIB.ORG

Challenges in the Computational Modeling of the Protein Structure—Activity Relationship

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Challenges in the Computational Modeling of the Protein Structure—Activity Relationship computation Review Challenges in the Computational Modeling of the Protein Structure—Activity Relationship Gabriel Del Río Department of Biochemistry and Structural Biology, Institute of Cellular Physiology, UNAM, Mexico City 04510, Mexico; [email protected] Abstract: Living organisms are composed of biopolymers (proteins, nucleic acids, carbohydrates and lipid polymers) that are used to keep or transmit information relevant to the state of these organisms at any given time. In these processes, proteins play a central role by displaying different activities required to keep or transmit this information. In this review, I present the current knowledge about the protein sequence–structure–activity relationship and the basis for modeling this relationship. Three representative predictors relevant to the modeling of this relationship are summarized to highlight areas that require further improvement and development. I will describe how a basic understanding of this relationship is fundamental in the development of new methods to design proteins, which represents an area of multiple applications in the areas of health and biotechnology. Keywords: protein structure; protein function; bijection 1. Proteins: Fundamental Polymers for Biological Systems Biological systems are composed of biopolymers, including nucleic acids, polymers Citation: Del Río, G. Challenges in (e.g., DNA, RNA), carbohydrates (e.g., starch, cellulose), lipids and proteins. All these the Computational Modeling of the biopolymers play critical roles in all the processes associated with life. Biopolymers are Protein Structure—Activity Relationship. Computation 2021, 9, 39. composed of monomers that are linked in a sequential way; for instance, proteins are https://doi.org/10.3390/ composed of amino acids, while nucleic acids are compose of nucleotides and polymers computation9040039 of carbohydrates are composed of different carbohydrate monomers (e.g., starch and cellulose are made of glucose mainly), and a similar situation occurs with lipids. Since each Academic Editor: Rainer Breitling biopolymer is composed of many different monomers, this situation gives rise to an enormous number of different biopolymer sequences. For instance, proteins are composed Received: 4 February 2021 of 20 different amino acids; hence, proteins that include up to 100 monomers of amino Accepted: 20 March 2021 acids will encompass 20100 different protein sequences. Published: 24 March 2021 The information about the state of the system is maintained by these biopolymers, and this is achieved by controlling the presence/activity of these biopolymers. Activity Publisher’s Note: MDPI stays neutral refers to the action that these biopolymers perform, either maintaining a particular morphol- with regard to jurisdictional claims in ogy, transporting cargo or altering the chemical structure of other molecules, among others. published maps and institutional affil- This activity is commonly referred to as function. In this review, I will use the term activity iations. to avoid confusion with the mathematical term function, which I will also use in this review. The flux of information in biological systems that involves all these biopolymers is depicted in Figure1, where proteins are proposed to play a central role. The central dogma in molecular biology [1] is depicted at the bottom of the figure, where the information flux Copyright: © 2021 by the author. goes from DNA to RNA (transcription) and from RNA to proteins (translation). The figure Licensee MDPI, Basel, Switzerland. includes bidirectional arrows to indicate that the information also goes from RNA to DNA This article is an open access article (retro-transcription) and from protein to RNA or DNA (epigenetics); the information flux distributed under the terms and may go from DNA to DNA, from RNA to RNA or from protein to protein, as indicated conditions of the Creative Commons by circular arrows. The other biopolymers (carbohydrates and lipids) are inserted into Attribution (CC BY) license (https:// this flux of information in the top part of the figure. It is relevant to note that proteins creativecommons.org/licenses/by/ participate in any of these fluxes, e.g., protein transcription factors are required for tran- 4.0/). Computation 2021, 9, 39. https://doi.org/10.3390/computation9040039 https://www.mdpi.com/journal/computation Computation 2021, 9, x FOR PEER REVIEW 2 of 11 Computation 2021, 9, 39 2 of 11 pids) are inserted into this flux of information in the top part of the figure. It is relevant to note that proteins participate in any of these fluxes, e.g., protein transcription factors are required for transcription, ribosomal proteins are involved in translation, and the syn- thesis/degradationscription, ribosomal of proteins carbohydrates are involved or lipids in translation, also requires and proteins. the synthesis/degradation of carbohydrates or lipids also requires proteins. Figure 1. Information flux among biopolymers. Figure 1. Information flux among biopolymers. Thus, proteins are central to life due to the different activities that they perform in biologicalThus, systems.proteins Inare consequence, central to life it due is expected to the different that knowledge activities of that the proteinthey perform contents in biologicalof living organisms systems. In may consequence, help to anticipate it is expected the capabilities that knowledge of such of systems. the protein For instance,contents ofhaving living access organisms to the may list ofhelp protein to anticipate sequences the in capabilities a tumor would of such provide systems. the For information instance, havingrequired access to anticipate to the list the of tumor’s protein malignancy; sequences in knowing a tumor the would proteins provide of viruses the information may antici- paterequired their to virulence. anticipate Due the totumor’s the advances malignancy; in DNA knowing sequencing the proteins techniques, of viruses having may access an- ticipateto protein their sequence virulence. data Due is relatively to the advances straightforward in DNA nowadays. sequencing However, techniques, knowledge having ac- of proteincess to sequencesprotein sequence is not enough data tois anticipate relatively the straightforward activities that proteins nowadays. may displayHowever, in knowledgebiological systems. of protein This sequences is because, is not at the enough most basicto anticipate level, protein the activities activity that depends protein ons maythe ability display of in proteins biological to interact systems. with This other is because, molecules; at the hence, most protein basic level, activity protein depends activity on dependsthe context on inthe which ability the of proteins protein isto located. interact Thiswith dependencyother molecules; of proteins hence, protein was originally activity dependsreferred toon as the gene context sharing, in which but, more the protein recently, is itlocated. has been This considered dependency a particular of proteins case was of originallya phenomena referred referred to as to gene as moonlighting sharing, but, [2 more]. For recently, instance, it when has been a protein considered referred a topar- as ticularthymidine case phosphorylaseof a phenomena is referred present to inside as moonlighting human cells, [2]. at For the cytoplasminstance, when of these a protein cells, referredit catalyzes to as the thymidine dephosphorylation phosphorylase of thymidine, is present among inside otherhuman nucleotides; cells, at the yet, cytoplasm when this of thesesame proteincells, it iscatalyzes expressed the outside dephosphorylation of cells, it stimulates of thymidine, cell growth among and other chemotaxis nucleotides; [3]. yet, whenFrom this the previoussame protein notion, is expressed it can then outside be surmised of cells, that it stimulates different contexts cell growth provide and proteinschemotaxis with [3]. different molecules to interact with, and this may be in turn translated into differentFrom protein the previous activities. notion, The it ways can then in which be surmised this differential that different binding contexts translates provide into proteinsdifferent with activities different highlights molecules another to interact feature ofwith, proteins: and this allostery may be [4 in]. Originallyturn translated proposed into differentby Monod protein and Jacob activities. in 1961, The the modelways in nowadays which this refers differential to conformational binding changestranslates in into any proteindifferent induced activities by hi bindingghlights to another other molecules, feature of which proteins: are requiredallostery for[4]. the Originally performance pro- posedof a particular by Monod activity and [Jacob5]. Take, in 1961, for instance, the model the nowadays case of hemoglobin refers to (Hb),conformational a protein changescarrying in two any oxygen protein molecules induced thatby binding are distributed to other frommolecules, blood which to tissues, are required which displays for the performanceseveral forms of of a allostery:particular (i)activity intrinsic, [5]. whenTake, for a single instance, oxygen the case molecule of hemoglobin binds to Hb (Hb), and a proteinthe binding carrying of the two second oxygen oxygen molecules molecule that is are facilitated distributed or (ii) from heterotrophic blood to tissues, ligands which (2,3- displaysbiphosphoglycerate, several forms protons, of allostery: carbon (i) dioxide, intrinsic, among when others) a single facilitate oxygen
Load more

Recommended publications
  • Arxiv:1911.09811V1 [Physics.Bio-Ph] 22 Nov 2019 Help to Associate a Folded Structure to a Protein Sequence
    Machine learning for protein folding and dynamics Frank Noé Department of Mathematics and Computer Science, Freie Universität Berlin, Arnimallee 6, 14195 Berlin, Germany Gianni De Fabritiis Computational Science Laboratory, Universitat Pompeu Fabra, Barcelona Biomedical Research Park (PRBB), Doctor Aiguader 88, 08003 Barcelona, Spain, and Institucio Catalana de Recerca i Estudis Avanats (ICREA), Passeig Lluis Companys 23, Barcelona 08010, Spain Cecilia Clementi Center for Theoretical Biological Physics, and Department of Chemistry, Rice University, 6100 Main Street, Houston, Texas 77005, United States Abstract Many aspects of the study of protein folding and dynamics have been affected by the recent advances in machine learning. Methods for the prediction of protein structures from their sequences are now heavily based on machine learning tools. The way simulations are performed to explore the energy landscape of protein systems is also changing as force-fields are started to be designed by means of machine learning methods. These methods are also used to extract the essential information from large simulation datasets and to enhance the sampling of rare events such as folding/unfolding transitions. While significant challenges still need to be tackled, we expect these methods to play an important role on the study of protein folding and dynamics in the near future. We discuss here the recent advances on all these fronts and the questions that need to be addressed for machine learning approaches to become mainstream in protein simulation. Introduction During the last couple of decades advances in artificial intelligence and machine learning have revolu- tionized many application areas such as image recognition and language translation.
    [Show full text]
  • Protein Contact Map Prediction Using Multiple Sequence Alignment
    bioRxiv preprint doi: https://doi.org/10.1101/2021.05.12.443740; this version posted May 13, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Protein contact map prediction using multiple sequence alignment dropout and consistency learning for sequences with less homologs Abstract The prediction of protein contact map needs enough normalized number of effective sequence (Nf) in multiple sequence alignment (MSA). When Nf is small, the predicted contact maps are often not satisfactory. To solve this problem, we randomly selected a small part of sequence homologs for proteins with large Nf to generate MSAs with small Nf. From these MSAs, input features were generated and were passed through a consistency learning network, aiming to get the same results when using the features generated from the MSA with large Nf. The results showed that this method effectively improves the prediction accuracy of protein contact maps with small Nf. 1. Introduction De novo protein structure prediction has been a long-standing challenge in computational biology. Since 2015, protein contact map assisted protein structure prediction has shown its great advantage and thus has been implemented in almost all cutting-edge de novo protein structure prediction tools, such as RaptorX1, I-TASSER2, AlphaFold13, trRosetta4 and etc5-8. Protein contact map is predicted by using features extracted from multiple sequences alignment (MSA). Such features include mutual information and direct-coupling analysis (DCA). Mutual information was often used in early stage studies to predict co-evolving residues pairs.9-11 However, mutual information includes both direct coupling pairs and indirect coupling pairs, while the latter are bioRxiv preprint doi: https://doi.org/10.1101/2021.05.12.443740; this version posted May 13, 2021.
    [Show full text]
  • Computational Protein Structure Prediction Using Deep Learning
    COMPUTATIONAL PROTEIN STRUCTURE PREDICTION USING DEEP LEARNING _______________________________________ A Dissertation presented to the Faculty of the Graduate School at the University of Missouri-Columbia _______________________________________________________ In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy _____________________________________________________ by ZHAOYU LI Professor Yi Shang, Dissertation Advisor May 2020 The undersigned, appointed by the dean of the Graduate School, have examined the dissertation entitled COMPUTATIONAL PROTEIN STRUCTURE PREDICTION USING DEEP LEARNING presented by Zhaoyu Li, a candidate for the degree of Doctor of Philosophy and hereby certify that, in their opinion, it is worthy of acceptance. Dr. Yi Shang Dr. Dong Xu Dr. Jianlin Cheng Dr. Ioan Kosztin DEDICATION To my wife Xian, my parents, and my son Ethan. ACKNOWLEDGEMENTS I would like to take the opportunity to thank the following people who have supported me through my PhD program. Without the support of them, this thesis would not have been possible. First, I would like to thank my advisor, Dr. Yi Shang. He has been my advisor since my master’s study. He opened the door of research for me, showed me instructions, guided me, supported me, encouraged me, and inspired me whenever I need help. It is a great pleasure to work with Dr. Shang together. I would also like to thank my committee members, Dr. Dong Xu, Dr. Jianlin Cheng, and Dr. Ioan Kosztin. They have been giving me constructive suggestions and advices to my dissertation and helping me solve problems in my research. Their support is essential, and I appreciate their time and efforts very much. I would also like to thank other lab mates.
    [Show full text]
  • Leveraging Novel Information Sources for Protein Structure Prediction
    Leveraging Novel Information Sources for Protein Structure Prediction vorgelegt von Dipl.-Biol. Michael Bohlke-Schneider, geboren Schneider geb. in Dortmund von der Fakultät IV — Elektrotechnik und Informatik der Technischen Universität Berlin zur Erlangung des akademischen Grades Doktor der Naturwissenschaften - Dr. rer. nat. - genehmigte Dissertation Promotionsausschuss: Vorsitzender: Prof. Dr. Klaus Obermayer 1. Gutachter: Prof. Dr. Oliver Brock 2. Gutachter: Prof. Dr. Juri Rappsilber 3. Gutachter: Prof. Dr. Jens Meiler Tag der wissenschaftlichen Aussprache: 22.12.2015 Berlin 2016 I would like to dedicate this thesis to my loving wife Nina and my mother Barbara. Acknowledgements Becoming a scientist was clearly the greatest challenge that I had to master in my life. The past five years changed my views and my life in a way that I never thought to be possible. Many people helped me during this journey. I am blessed with having many good friends and I cannot possibly list all the people that I am grateful for. First, I want to thank my adviser, Oliver Brock. Oliver is the sharpest mind I have ever met and his ability to think incredibly clearly has been very inspiring. In addition, Oliver has redefined my standard for scientific rigor and what it means to master all skills that a scientist should have. But his most important lesson was to never be satisfied, to constantly challenge the status quo, and to change the world for the better. This lesson will surely shape my further life, no matter where it will take me. I would also like to thank the members of my committee, Juri Rappsilber and Jens Meiler.
    [Show full text]
  • Ensembling Multiple Raw Coevolutionary Features with Deep Residual Neural Networks for Contact‐Map Prediction in CASP13
    Received: 17 April 2019 Revised: 20 July 2019 Accepted: 8 August 2019 DOI: 10.1002/prot.25798 RESEARCH ARTICLE Ensembling multiple raw coevolutionary features with deep residual neural networks for contact-map prediction in CASP13 Yang Li1,2 | Chengxin Zhang2 | Eric W. Bell2 | Dong-Jun Yu1,2 | Yang Zhang2 1School of Computer Science and Engineering, Nanjing University of Science and Technology, Abstract Nanjing, China We report the results of residue-residue contact prediction of a new pipeline built 2 Department of Computational Medicine and purely on the learning of coevolutionary features in the CASP13 experiment. For a Bioinformatics, University of Michigan, Ann Arbor, Michigan, USA query sequence, the pipeline starts with the collection of multiple sequence alignments (MSAs) from multiple genome and metagenome sequence databases using two comple- Correspondence Yang Zhang, Department of Computational mentary Hidden Markov Model (HMM)-based searching tools. Three profile matrices, Medicine and Bioinformatics, University of built on covariance, precision, and pseudolikelihood maximization respectively, are then Michigan, Ann Arbor, MI 48109. Email: [email protected] created from the MSAs, which are used as the input features of a deep residual con- volutional neural network architecture for contact-map training and prediction. Two Dong-Jun Yu, School of Computer Science and Engineering, Nanjing University of Science and ensembling strategies have been proposed to integrate the matrix features through Technology, Xiaolingwei 200, Nanjing
    [Show full text]
  • Direct Information Reweighted by Contact Templates: Improved RNA Contact Prediction by Combining Structural Features
    Direct Information Reweighted by Contact Templates: Improved RNA Contact Prediction by Combining Structural Features Yiren Jian2, Chen Zeng2*, Yunjie Zhao1* Author affiliation: 1Institute of Biophysics and Department of Physics, Central China Normal University, Wuhan 430079, China 2Department of Physics, The George Washington University, Washington, DC 20052, USA *Corresponding author Chen Zeng Email: [email protected] Yunjie Zhao Email: [email protected] Abstract It is acknowledged that co-evolutionary nucleotide-nucleotide interactions are essential for RNA structures and functions. Currently, direct coupling analysis (DCA) infers nucleotide contacts in a sequence from its homologous sequence alignment across different species. DCA and similar approaches that use sequence information alone usually yield a low accuracy, especially when the available homologous sequences are limited. Here we present a new method that incorporates a Restricted Boltzmann Machine (RBM) to augment the information on sequence co-variations with structural patterns in contact inference. We thus name our method DIRECT that stands for Direct Information REweighted by Contact Templates. Benchmark tests demonstrate that DIRECT produces a substantial enhancement of 13% in accuracy on average for contact prediction in comparison to the traditional DCA. These results suggest that DIRECT could be used for improving predictions of RNA tertiary structures and functions. The source codes and dataset of DIRECT are available at http:// http://zhao.phy.ccnu.edu.cn:8122/DIRECT/index.html. Introduction RNA molecules play critical roles in various biological processes (1-8). There remain various challenges in determination of RNA structure through experiments to get a comprehensive understanding of structure-function relation for RNAs (9).
    [Show full text]
  • Machine Learning Methods for Protein Structure Prediction Jianlin Cheng, Allison N
    IEEE REVIEWS IN BIOMEDICAL ENGINEERING, VOL. 1, 2008 41 Machine Learning Methods for Protein Structure Prediction Jianlin Cheng, Allison N. Tegge, Member, IEEE, and Pierre Baldi, Senior Member, IEEE Methodological Review Abstract—Machine learning methods are widely used in bioin- formatics and computational and systems biology. Here, we review the development of machine learning methods for protein structure prediction, one of the most fundamental problems in structural biology and bioinformatics. Protein structure predic- tion is such a complex problem that it is often decomposed and attacked at four different levels: 1-D prediction of structural features along the primary sequence of amino acids; 2-D predic- tion of spatial relationships between amino acids; 3-D prediction Fig. 1. Protein sequence-structure-function relationship. A protein is a linear of the tertiary structure of a protein; and 4-D prediction of the polypeptide chain composed of 20 different kinds of amino acids represented quaternary structure of a multiprotein complex. A diverse set of by a sequence of letters (left). It folds into a tertiary (3-D) structure (middle) both supervised and unsupervised machine learning methods has composed of three kinds of local secondary structure elements (helix – red; beta- strand – yellow; loop – green). The protein with its native 3-D structure can carry been applied over the years to tackle these problems and has sig- out several biological functions in the cell (right). nificantly contributed to advancing the state-of-the-art of protein structure prediction. In this paper, we review the development and application of hidden Markov models, neural networks, support vector machines, Bayesian methods, and clustering methods in structure elements are further packed to form a tertiary structure 1-D, 2-D, 3-D, and 4-D protein structure predictions.
    [Show full text]
  • Protein Structures
    Protein Structure Determination Dr. Fayyaz ul Amir Afsar Minhas PIEAS Biomedical Informatics Research Lab Department of Computer and Information Sciences Pakistan Institute of Engineering & Applied Sciences PO Nilore, Islamabad, Pakistan http://faculty.pieas.edu.pk/fayyaz/ CIS 529: Bioinformatics PIEAS Biomedical Informatics Research Lab Protein Structure • How is the structure of a protein determined? – Sequencing – Tertiary Structure Determination • X-ray crystallography • NMR Spectroscopy • Cryo-Electron microscopy • Structural Alignment • Structure Prediction – Secondary and Super-Secondary Structure and Properties Prediction – Tertiary Structure Prediction – Structural Dynamics • Protein Interactions – Interaction Prediction – Binding site prediction • Docking CIS 529: Bioinformatics PIEAS Biomedical Informatics Research Lab 2 Protein Sequencing • Protein Sequencer – They work by tagging and removing one amino acid at a time, which is analysed and identified. This is done repetitively for the whole polypeptide, until the whole sequence is established. – This method has generally been replaced by nucleic acid technology, and it is often easier to identify A Beckman-Coulter Porton the sequence of a protein by looking at the DNA LF3000G protein sequencing that codes for it. machine • Mass-Spectrometry • Edman Degradation https://en.wikipedia.org/wiki/Protein_sequencing CIS 529: Bioinformatics PIEAS Biomedical Informatics Research Lab 3 X-ray Crystallography • 1960 John Kendrew determined the first structure of a protein using X-ray diffraction
    [Show full text]
  • Deep Learning-Based Advances in Protein Structure Prediction
    International Journal of Molecular Sciences Review Deep Learning-Based Advances in Protein Structure Prediction Subash C. Pakhrin 1,†, Bikash Shrestha 2,†, Badri Adhikari 2,* and Dukka B. KC 1,* 1 Department of Electrical Engineering and Computer Science, Wichita State University, Wichita, KS 67260, USA; [email protected] 2 Department of Computer Science, University of Missouri-St. Louis, St. Louis, MO 63121, USA; [email protected] * Correspondence: [email protected] (B.A.); [email protected] (D.B.K.) † Both the authors should be considered equal first authors. Abstract: Obtaining an accurate description of protein structure is a fundamental step toward under- standing the underpinning of biology. Although recent advances in experimental approaches have greatly enhanced our capabilities to experimentally determine protein structures, the gap between the number of protein sequences and known protein structures is ever increasing. Computational protein structure prediction is one of the ways to fill this gap. Recently, the protein structure prediction field has witnessed a lot of advances due to Deep Learning (DL)-based approaches as evidenced by the suc- cess of AlphaFold2 in the most recent Critical Assessment of protein Structure Prediction (CASP14). In this article, we highlight important milestones and progresses in the field of protein structure prediction due to DL-based methods as observed in CASP experiments. We describe advances in various steps of protein structure prediction pipeline viz. protein contact map prediction, protein distogram prediction, protein real-valued distance prediction, and Quality Assessment/refinement. We also highlight some end-to-end DL-based approaches for protein structure prediction approaches. Additionally, as there have been some recent DL-based advances in protein structure determination Citation: Pakhrin, S.C.; Shrestha, B.; using Cryo-Electron (Cryo-EM) microscopy based, we also highlight some of the important progress Adhikari, B.; KC, D.B.
    [Show full text]
  • Protein Contact Map Denoising Using Generative Adversarial Networks
    bioRxiv preprint doi: https://doi.org/10.1101/2020.06.26.174300; this version posted June 27, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Protein Contact Map Denoising Using Generative Adversarial Networks Sai Raghavendra Maddhuri Venkata Subramaniya1, Genki Terashi2, Aashish Jain1, Yuki Kagaya3, and Daisuke Kihara*2,1 1 Department of Computer Science, Purdue University, West Lafayette, IN, 47907, USA 2 Department of Biological Sciences, Purdue University, West Lafayette, IN, 47907, USA 3 Graduate School of Information Sciences, Tohoku University, Japan * To whom correspondence should be addressed E-mail: [email protected] Tel: 1-765-496-2284 Fax: 1-765-496-1189 1 bioRxiv preprint doi: https://doi.org/10.1101/2020.06.26.174300; this version posted June 27, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. ABSTRACT Protein residue-residue contact prediction from protein sequence information has undergone substantial improvement in the past few years, which has made it a critical driving force for building correct protein tertiary structure models. Improving accuracy of contact predictions has, therefore, become the forefront of protein structure prediction. Here, we show a novel contact map denoising method, ContactGAN, which uses Generative Adversarial Networks (GAN) to refine predicted protein contact maps.
    [Show full text]
  • Homology Modeling in the Time of Collective and Artificial Intelligence
    Computational and Structural Biotechnology Journal 18 (2020) 3494–3506 journal homepage: www.elsevier.com/locate/csbj Homology modeling in the time of collective and artificial intelligence ⇑ Tareq Hameduh a, Yazan Haddad a,b, Vojtech Adam a,b, Zbynek Heger a,b, a Department of Chemistry and Biochemistry, Mendel University in Brno, Zemedelska 1, CZ-613 00 Brno, Czech Republic b Central European Institute of Technology, Brno University of Technology, Purkynova 656/123, 612 00 Brno, Czech Republic article info abstract Article history: Homology modeling is a method for building protein 3D structures using protein primary sequence and Received 14 August 2020 utilizing prior knowledge gained from structural similarities with other proteins. The homology modeling Received in revised form 4 November 2020 process is done in sequential steps where sequence/structure alignment is optimized, then a backbone is Accepted 4 November 2020 built and later, side-chains are added. Once the low-homology loops are modeled, the whole 3D structure Available online 14 November 2020 is optimized and validated. In the past three decades, a few collective and collaborative initiatives allowed for continuous progress in both homology and ab initio modeling. Critical Assessment of protein Keywords: Structure Prediction (CASP) is a worldwide community experiment that has historically recorded the pro- Homology modeling gress in this field. Folding@Home and Rosetta@Home are examples of crowd-sourcing initiatives where Machine learning Protein 3D structure the community is sharing computational resources, whereas RosettaCommons is an example of an initia- Structural bioinformatics tive where a community is sharing a codebase for the development of computational algorithms.
    [Show full text]
  • And Contact-Based Protein Structure Prediction
    International Journal of Molecular Sciences Review Recent Applications of Deep Learning Methods on Evolution- and Contact-Based Protein Structure Prediction Donghyuk Suh 1 , Jai Woo Lee 1 , Sun Choi 1 and Yoonji Lee 2,* 1 Global AI Drug Discovery Center, School of Pharmaceutical Sciences, College of Pharmacy and Graduate, Ewha Womans University, Seoul 03760, Korea; [email protected] (D.S.); [email protected] (J.W.L.); [email protected] (S.C.) 2 College of Pharmacy, Chung-Ang University, Seoul 06974, Korea * Correspondence: [email protected] Abstract: The new advances in deep learning methods have influenced many aspects of scientific research, including the study of the protein system. The prediction of proteins’ 3D structural compo- nents is now heavily dependent on machine learning techniques that interpret how protein sequences and their homology govern the inter-residue contacts and structural organization. Especially, meth- ods employing deep neural networks have had a significant impact on recent CASP13 and CASP14 competition. Here, we explore the recent applications of deep learning methods in the protein structure prediction area. We also look at the potential opportunities for deep learning methods to identify unknown protein structures and functions to be discovered and help guide drug–target interactions. Although significant problems still need to be addressed, we expect these techniques in the near future to play crucial roles in protein structural bioinformatics as well as in drug discovery. Keywords: structural bioinformatics; deep learning; protein sequence homology; 3D structure of Citation: Suh, D.; Lee, J.W.; Choi, S.; Lee, Y. Recent Applications of Deep proteins; drug discovery Learning Methods on Evolution- and Contact-Based Protein Structure Prediction.
    [Show full text]