DMFold: Protein complex structure prediction based on DeepMSA2

Online Services

●I-TASSER ●I-TASSER-MTD ●C-I-TASSER ●CR-I-TASSER ●QUARK ●C-QUARK ●LOMETS ●MUSTER ●CEthreader ●SEGMER ●DeepFold ●DeepFoldRNA ●FoldDesign ●COFACTOR ●COACH ●MetaGO ●TripletGO ●IonCom ●FG-MD ●ModRefiner ●REMO ●DEMO ●DEMO-EM ●DMFold ●SPRING ●COTH ●Threpp ●PEPPI ●BSpred ●ANGLOR ●EDock ●BSP-SLIM ●SAXSTER ●FUpred ●ThreaDom ●ThreaDomEx ●EvoDesign ●BindProf ●BindProfX ●SSIPe ●GPCR-I-TASSER ●MAGELLAN ●ResQ ●STRUM ●DAMpred

●TM-score ●TM-align ●US-align ●MM-align ●RNA-align ●NW-align ●LS-align ●EDTSurf ●MVP ●MVP-Fit ●SPICKER ●HAAD ●PSSpred ●3DRobot ●MR-REX ●I-TASSER-MR ●SVMSEQ ●NeBcon ●ResPRE ●TripletRes ●DeepPotential ●WDL-RF ●ATPbind ●DockRMSD ●DeepMSA ●FASPR ●EM-Refiner ●GPU-I-TASSER

●BioLiP ●E. coli ●GLASS ●GPCR-HGmod ●GPCR-RD ●GPCR-EXP ●Tara-3D ●TM-fold ●DECOYS ●POTENTIAL ●RW/RWplus ●EvoEF ●HPSF ●THE-DB ●ADDRESS ●Alpaca-Antibody ●CASP7 ●CASP8 ●CASP9 ●CASP10 ●CASP11 ●CASP12 ●CASP13 ●CASP14

DMFold (also known as DMFold-Multimer) is a deep learning-based approach to protein complex structure and function prediction built on deep multiple sequence alignments (MSAs). The core of the pipeline is the integration of DeepMSA2 with the modified structure module of AlphaFold2. Starting from a set of query sequences, DMFold first creates deep monomeric MSAs using an iterative search procedure through multiple whole-genome (Uniclust30 and UniRef90) and metagenome (Metaclust, BFD, Mgnify, TaraDB, MetaSourceDB and JGIclust) databases, where multimeric MSAs are then constructed by pairing the monomeric MSAs based on species annotations. Next, complex structure models are predicted by feeding the multimetic MSAs into the structural modules of AlphaFold2-Multimer, where funtional annotations, including Gene Ontology, Enzyme Commission and Ligand Binding Sites, are generated by COFACTOR2 and US-align based on the top DMFold structure models. DMFold participated (as "Zheng") in CASP15 and ranked as the No. 1 method for protein-protein complex structure prediction, with accuracy significantly higher than the state-of-the-art AlphaFold2 program (i.e., "NBIS-AF2-multimer" in CASP15).

Although DMFold focuses on multi-chain protein complexes, it also accepts single-chain monomer sequences (DMFold-Monomer pipeline). The server is freely accessible to all users, including commercial ones. Please report problems and questions at Zhang Lab Server Forum.

Notice: During the CASP season (04/23/2024-09/01/2024), DMFold server will decrease the allowed maximal size of protein complex to 2000. After CASP season, we will consider increase it again!

If you have a large amount of targets want to submit to DMFold server, please consider seeking collaboration with our lab, you can contact zhengwei@umich.edu.

The outputs of the DMFold server include (see Example outputs for complex and Example outputs for monomer):

Best MSAs from DeepMSA2.
Contact and distance maps predicted by DMFold.
Individual chain-level structure and function predictions.
Full-chain structural models built by DMFold.
Residue-level pLDDT score of each individual chain.
Top ten PDB structures closest the target models identified by US-align/TM-align.
Gene Ontology (BP, CC and MF) annotations derived from top templates (for complexes), or predicted by COFACTOR2 (for monomers).
Enzyme Commission annotations derived from top templates (for complexes), or predicted by COFACTOR2 (for monomers).
Ligand Binding Sites derived from top templates (for complexes), or predicted by COFACTOR2 (for monomers).

[Example outputs for complex] [Example outputs for monomer] [Benchmark Dataset] [Standalone package] [Human Proteome] [Check Previous Jobs] [Help] [Forum]

Online server

Copy and paste your sequence (in FASTA format) below (The server accepts both monomer and complex sequences with length in [30, 1200 AA] for monomers, and in [30, 2000 AA] for complexes with < 10 chains):
[Example inputs of monomer] [Example inputs of complex]

Or upload sequences from your local computer:

Email: (mandatory, where results will be sent to)

ID: (optional, your given name of the protein)

Advanced options

MSA generation method:
Utilizing Uniclust30, Uniref90 and Metaclust databases (fast).
Utilizing Uniclust30, Uniref90, Metaclust, MGnify and BFD databases (medium).
Utilizing Uniclust30, Uniref90, Metaclust, MGnify, BFD, TaraDB, MetaSourceDB and JGIclust databases (slow).

Function prediction:
Predict protein function based on structure models (additional one to five days may be applied based on the size).

DMFold News:

2024/03/22:

DMFold (v1.2)

2024/02/09:

2024/01/02:

2023/01/01:

CASP15

References:

Wei Zheng, Qiqige Wuyun, Yang Li, Chengxin Zhang, P Lydia Freddolino, Yang Zhang. Improving deep learning protein monomer and complex structure prediction using DeepMSA2 with huge metagenomics data. Nature Methods, (2024). https://doi.org/10.1038/s41592-023-02130-4.
Wei Zheng, Quancheng Liu, Qiqige Wuyun, P. Lydia Freddolino, Yang Zhang. DMFold: A deep learning platform for protein complex structure and function predictions based on DeepMSA2. In preparation.
Wei Zheng, Qiqige Wuyun, Peter L Freddolino, Yang Zhang. Integrating deep learning, threading alignments, and a multi-MSA strategy for high-quality protein monomer and complex structure prediction in CASP15. 1-20. Proteins. (2023). doi:10.1002/prot.26585.