WIXLIB

CRITICAL ASSESSMENT OF TECHNIQUESFOR PROTEIN STRUCTURE PREDICTION14ABSTRACT BOOKFourteenth roundMay-September 20201

TABLE OF CONTENTS191227 . 13iPhord: A protein structure prediction system based on deep learning . 133DCNN prof . 16Single model quality assessment using 3DCNN with profile-based features . 16A2I2Prot . 17Fusion of sequence embedding and sequence alignments for protein contact predictions . 17AILON . 19Protein tertiary structure prediction driven by deep neural network and cmFinder from its amino acidsequence . 19AIR . 21AIR: An artificial intelligence-based protocol for protein structure refinement using multi-objectiveparticle swarm optimization . 21AlphaFold2 . 22High Accuracy Protein Structure Prediction Using Deep Learning . 22angleQA . 25AngleQA: protein single-model quality assessment based on torsion angles . 25AP 1 . 27AP 1 structure predictions in CASP14 . 27BAKER-experimental (Assembly). 28Protein oligomer structure predictions guided by predicted inter-chain contacts . 28BAKER-ROSETTASERVER, BAKER (TS) . 30Protein structure prediction guided by predicted inter-residue geometries . 30BAKER-ROSETTASERVER, BAKER-experimental (EMA) . 32Estimation of Model Quality via Deep Residual Learning . 32BAKER, BAKER-experimental (Refinement) . 34Model refinement guided by an interplay between Deep-learning and Rosetta. 34Bates BMM . 36Protein fold construction and complex assembly by employing particle swarm optimization. 36Bhattacharya . 38Protein tertiary structure prediction by Bhattacharya group in CASP14 . 381

Bhattacharya-QDeep, Bhattacharya-QDeepU, Bhattacharya-Server . 39Protein model accuracy estimation by Bhattacharya groups in CASP14. 39Bhattacharya, Bhattacharya-Server . 40Protein structure refinement by Bhattacharya groups in CASP14 . 40Bioinsilico sbi . 42Three-dimensional prediction of proteins using a collection of sMotifs . 42Bioinsilico sbi . 44Protein contact predictions using a reduced alphabet and direct-coupling analysis . 44Bioinsilico sbi, Bioinsilico sbi PAIR. 46Assessing the quality of protein structural models using split-statistical potentials . 46BrainFold . 48Contact Pair Prediction Using a Deep Neural Net. 48CAO-QA1 (High Accuracy Modeling) . 51Collaborative protein structure prediction with deep learning based de novo prediction and modelselection . 51CAO-QA1(Accuracy Estimation) . 54AngularQA: Protein Model Quality Assessment with LSTM Networks. 54CAO-SERVER(Accuracy Estimation) . 57TopQA: a topological representation for single-model protein quality assessment with machinelearning . 57CAO-SERVER(Topology) . 60de novo protein structure prediction using stepwise fragment sampling with. 60contact prediction and model selection based on deep learning techniques. 60ClusPro . 62Hybrid ClusPro server in 2020 CASP/CAPRI rounds . 62CMH1971 . 64Structure Prediction, Quality Assessment and Contact Prediction by EMAP CLUST . 64CUTSP . 65Morphing semi-supervised protein structures predicted using distance and torsion representationswith deep graph ranking . 65DATE . 672

Template-based Structure Prediction and Interresidue Distances and Orientations Prediction-basedStructure Prediction . 67DeepML . 69Quality Assesment of Protein Models using Graph Convolutional Networks . 69DeepMUSICS. 69A novel deep learning framework for protein structure prediction . 69DeepPotential. 72Learning deep statistical potentials for protein folding. 72DELCLAB . 74DellaCorteLab . 76Refinement with Improved Restrained Molecular Dynamics . 76DellaCorteLab . 78De novo structure prediction with deep learning and molecular mechanics simulation. 78DellaCorteLab . 80ProSPr: Protein Structure Prediction via Inter-Residue Distances. 80DESTINI . 82Deep-learning the protein folding code for structure prediction and sequence comparison . 82DMP2. 84Tertiary structure and distance predictions with DMPfold2 . 84E2E. 86E2E: Towards an end to end structure prediction pipeline. . 86edmc pf . 88Protein folding from contact maps using Euclidean distance matrix completion . 88EDN . 89Protein model quality assessment solely based on the structure of individual models . 89Elofsson . 90Template based and free docking in CASP14 . 90Elofsson . 94Advanced Deep-Learning applications on CASP14 protein models quality assessment . 94EMAP CHAE . 973

Protein model quality assessment method EMAP in CASP14 . 97FALCON-DeepFolder . 98CopulaNet: Learning residue co-evolution directly from multiple sequence alignment for proteinstructure prediction . 98FALCON-geom. 100Improving protein tertiary structure construction through reducing inconsistency from the predictedinter-residue distances . 100FALCON-TBM . 102FALCON-TBM: Accurate protein threading through learning structure alignment using neural network. 102FEIG and FEIG-R1,2,3 (TS) . 104Protein structure prediction via refinement of CASP server models . 104FEIG-S (refinement) . 104Protein Model Refinement via Molecular Dynamics Simulations with an Improved Structure SamplingProtocol and Multiple Alternative Models . 104FEIG-S (TS) . 108High Accuracy Protein Structure Prediction via Contact Prediction and Physics-based Refinement. 108Fernandez-Recio . 111Assembly prediction in CASP14 with pyDock ab initio docking and scoring . 111Frishman Group . 114DeepHelicon: Accurate prediction of inter-helical residue contacts in transmembrane proteins byresidual neural networks . 114GAPF LNCC . 116GAPF LNCC SERVER: an automated template based de novo protein structure prediction methodwith a multiple minima genetic algorithm. 116GLoSA . 119Refinement Using Molecular Dynamics with Restraints Derived from Binding Site Templates . 119graph-sh . 120Spherical convolutions on molecular graphs for protein model quality assessment . 120HMS-Casper . 122Recurrent geometric networks using Frenet-Serret geometry and latent residue representations . 122Huang . 1244

Template-based and Contact-assisted Docking for Homo-oligomeric Targets . 124ICOS . 127Contact map prediction using a residual and fully convolutional neural network . 127IntFOLD6. 130Fully Automated Prediction of Protein Tertiary Structures with Local Model Quality Scores Using theIntFOLD6 Server . 130JLU Comp Struct Bio . 133Molecular free energy optimization on a computational graph . 133Jones-UCL . 135Manually curated tertiary structure and distance predictions using DMPfold2 . 135Kiharalab . 137Distance Prediction, Structure Prediction, Refinement, Quality Assessment, and Protein Docking inKiharaLab . 137Kiharalab Contact . 140Protein Distance and Contact Prediction with Deep Learning . 140Kiharalab Refine . 142Integrative modeling for protein structure refinement using Molecular Dynamics with flat-bottomharmonic restraints, enhanced sampling and ROSETTA iterative hybridize. 142Kiharalab Z Server . 145Automatic Prediction of Protein Structure by Deep Learning and Rank Aggregation. 145Kozakov-Vajda . 148Template-assisted docking and docking of protein models. . 148LamoreuxLab . 150Modeling and assessment of CASP14 Targets using 3DCNN and template-based methods . 150Laufer ros . 151Protein Folding with MELD Molecular Dynamics. 151LAW, MASS . 154Predicting protein residue-residue contacts and tertiary structures using deep networks with varyingdilation rates . 154LAW, MASS (QA) . 156Protein Single-Model Accuracy Estimation Using Graph and Residual Neural Networks . 1565

McGuffin . 158Manual Prediction of Protein Tertiary and Quaternary Structures and 3D Model Refinement . 158MESHI . 162The human MESHI group in CASP14 . 162MESHI consensus . 163The MESHI concensus server for estimation of model accuracy. 163MESHI EMA. 165The MESHI EMA server for estimation of model accuracy . 165MESHI server . 167The MESHI-server pipeline for estimation of model accuracy and template based structure prediction. 167ModFOLD8, ModFOLD8 cor, ModFOLD8 rank . 169Automated 3D Model Quality Assessment using the ModFOLD8 Server. 169ModFOLDclust2 . 172Automated 3D Model Quality Assessment using ModFOLDclust2. 172MUFOLD . 174MUFOLD REFINE: Iterative Protein Structure Refinement Using Potential Functions based on DistanceDistributions . 174MUFOLD H . 176MUFOLD HUMAN: Protein Structure Prediction Using . 176Potential Functions based on Distance Distributions . 176MUfoldQA G . 178MUfoldQA G: A New Multi-Model QA Method Based on Machine Learning . 178MUfoldQA X . 180MufoldQA X: A New Quasi-Single-Model QA Method Combining Template Based and Deep LearningBased Features . 180MULTICOM (TS) . 182CASP14 Tertiary Structure Prediction by MULTICOM Human Group . 182MULTICOM-AI (TS) . 183Prediction of Protein Interchain Contacts and Complex Structures in CASP14-CAPRI Experiment . 183MULTICOM-CLUSTER, MULTICOM-CONSTRUCT, MULTICOM-AI (QA) . 1856

Protein model quality assessment with deep learning and residue-residue contact and distancepredictions . 185MULTICOM-CONSTRUCT, MULTICOM-CLUSTER, MULTICOM-DIST, MULTICOMHYBRID, MULTICOM-DEEP (TS) . 187CASP14 Protein Tertiary Structure Prediction by MULTICOM Server Predictors . 187MULTICOM-DEEP, MULTICOM-DIST, MULTICOM-HYBRID (QA) . 189Improving protein single-model and consensus quality assessment using inter-residue distanceprediction and deep learning . 189MULTICOM, MULTICOM-DEEP, MULTICOM-CONSTRUCT, MULTICOM-DIST,MULTICOM-AI, MULTICOM-HYBRID, MULTICOM-CLUSTER (RR). 192Prediction of protein inter-residue distance and contacts with deep learning . 192netris . 195netris: Contact Prediction through Network Inference Methods. 195NOVA. 197NOVA: protein structure prediction with structural information and deep learning-based foldingframework. 197Ornate . 199Protein model quality assessment using 3D oriented convolutional neural network Ornate . 199PBuild . 201A Rust-based Protein Tertiary Structure Builder . 201PerezLab Gators . 203MELDxMD: incorporating ML derived distograms and heuristics into simulations . 203PerillaGroup . 205High throughput structural refinement for large-scale MD simulations . 205Pharmulator . 208Pharmulator Structure Prediction Module: Angle-Based Structure Prediction using Bidirectional LSTMand Random Omega Generation . 208PrayogRealDistance . 211Real-valued protein distance prediction. 211PreferredFold . 213The PreferredFold pipeline: a top-down approach to predicting protein structure . 213Protein-blacksmith . 2167

Protein structure refinement: how frustration analysis assists principal component guided simulations. 216QMEANDisCo . 218QMEANDisCo – distance constraints applied on model quality estimation . 218QUARK. 220Protein 3D Structure Prediction by D-QUARK in CASP14 . 220RaptorX . 223Improved Protein Contact and Structure Prediction by Deep Learning . 223RaptorX-QA . 226Improved protein model quality assessment by integrating sequential and pairwise features usingdeep learning . 226RBO-PSP-CP . 228Protein Structure and Distance Prediction by RBO in CASP 14 . 228Risoluto . 230Holistic Approach to Integrate Template-based and Template-free Modeling . 230ropius0 . 233ROPIUS0: Restraint-Oriented Protocol for Inference and Understanding of protein Structures . 233ropius0QA . 236Ranking CASP-hosted server predictions by the accuracy estimation module of the ROPIUS0 protocol. 236Rostlab . 238EMBER: Predicting inter-residue distances using novel sequence embeddings . 238SASHAN, UOSHAN, KUHHAN (QA, TS) . 240Estimation of protein model accuracy by the predicted features, torsion potential, templates andclustering algorithm in CASP14 . 240SASHAN, UOSHAN, KUHHAN (RR) . 242Prediction of residue conact and distance using deep learning in CASP14 . 242SBROD . 244Evaluation of coarse-grained scoring function SBROD in the QA category of CASP14 . 244Seok-assembly, Seok (Assembly) . 246Modeling oligomeric proteins by GALAXY in CASP14 . 2468