List of protein structure prediction software


[Image:Protein-structure.png|right|thumb|Constituent amino-acids can be analyzed to predict secondary, tertiary and quaternary protein structure.]
This list of protein structure prediction software summarizes notable used software tools in protein structure prediction, including homology modeling, protein threading, ab initio methods, secondary structure prediction, and transmembrane helix and signal peptide prediction.

Software list

Below is a list which separates programs according to the method used for structure prediction.

Homology modeling

NameMethodDescriptionRelease date
IntFOLDA unified interface for: Tertiary structure prediction/3D modelling, 3D model quality assessment, Intrinsic disorder prediction, Domain prediction, Prediction of protein-ligand binding residuesAutomated webserver and some downloadable programs-
RaptorXremote homology detection, protein 3D modeling, binding site predictionAutomated webserver and Downloadable program-
Biskitwraps external programs into automated workflowBLAST search, T-Coffee alignment, and MODELLER construction-
ESyPred3DTemplate detection, alignment, 3D modelingAutomated webserver-
FoldXEnergy calculations and protein designDownloadable program-
Phyre, Phyre2Remote template detection, alignment, 3D modeling, multi-templates, ab initioWebserver with job manager, automatically updated fold library, genome searching and other facilities-
HHpredTemplate detection, alignment, 3D modelingInteractive webserver with help facility-
MODELLERSatisfaction of spatial restraintsStandalone program mainly in Fortran and Python-
CONFOLDSatisfaction of contact and distance restraintsStandalone program mainly in Fortran and Perl-
Molecular Operating Environment Template identification, use of multiple templates and accounting for other environments, loop modelling, rotamer libraries for sidechain conformations, relaxation using MM forcefields.Proprietary platform, supported on Windows, Linux and Mac-
RobettaRosetta homology modeling and ab initio fragment assembly with Ginzu domain predictionWebserver-
BHAGEERATH-HCombination of ab initio folding and homology methodsProtein tertiary structure predictions-
Swiss-modelLocal similarity/fragment assemblyAutomated webserver -
YasaraDetection of templates, alignment, modeling incl. ligands and oligomers, hybridization of model fragmentsGraphical interface or text mode -
AWSEM-SuiteMolecular dynamics simulation based on template-guided, coevolutionary-enhanced optimized folding landscapesAutomated webserver-

Threading and fold recognition

NameMethodDescriptionRelease date
IntFOLDA unified interface for: Tertiary structure prediction/3D modelling, 3D model quality assessment, Intrinsic disorder prediction, Domain prediction, Prediction of protein-ligand binding residuesAutomated webserver and some downloadable programs-
RaptorXRemote template detection, single-template and multi-template threading, totally different from and much better than the old program RAPTOR designed by the same groupWebserver with job manager, automatically updated fold library-
HHpredTemplate detection, alignment, 3D modelingInteractive webserver with help facility-
Phyre, Phyre2Remote template detection, alignment, 3D modeling, multi-templates, ab initioWebserver with job manager, automatically updated fold library, genome searching and other facilities-
I-TASSERThreading fragment structure reassemblyOn-line server for protein modeling https://zhanggroup.org/I-TASSER/-

''Ab initio'' structure prediction

NameMethodDescriptionRelease date
trRosettatrRosetta is an algorithm for fast and accurate protein structure prediction. It supports single-sequence structure prediction with trRosettaX-Single.Webserver and source codes available at: https://yanglab.qd.sdu.edu.cn/trRosetta/-
ROBETTARosetta homology modeling and ab initio fragment assembly with Ginzu domain predictionWebserver-
Rosetta@homeDistributed-computing implementation of Rosetta algorithmDownloadable program-
AbaloneMolecular Dynamics foldingProgram-
C-QUARKC-QUARK is a method for ab initio protein structure prediction. Based on deep-learning based contact-map predictions into the fragment assembly simulations.Webserver https://zhanggroup.org/C-QUARK/-
AlphaFold2An end-to-end deep learning framework for protein structure predictionWebserver and downloadable program-
D-I-TASSERThe deep learning-based I-TASSER program reportedly outperforms both AlphaFold2 and AlphaFold3Webserver https://zhanggroup.org/D-I-TASSER/-

Secondary structure prediction

Detailed list of programs can be found at List of protein secondary structure prediction programs