HPmod (Human Proteome Model) database is a repository for human protein structure and function that automatically predicted by the state-of-the-art algorithm from Yang Zhang Lab. Protein structure models are predicted by D-I-TASSER, and the protein functions, including Gene Ontology (GO), Enzyme Commission (EC), and ligand-binding sites, are predicted by COFACTOR. The database contains 19,512 proteins collected from Uniprot, which can be classified as 12,236 single-domain proteins and 7,276 multi-domain proteins. (>>more about the HPmod database)
Browse all proteins generated by D-I-TASSER in one page, loading slow.
Browse proteins generated by D-I-TASSER in the seperated pages, loading fast.
Browse 1,689 proteins generated by D-I-TASSER-AF2 in one pages, loading fast.
Browse 1,689 proteins generated by D-I-TASSER-AF2 in the seperated pages, loading fast.
The current database contains 19,512 entries.
Apr 30, 2022:
76 models for human complete genome proteins recently reported by science journal are generated by D-I-TASSER-AF2 modeling.
Apr 20, 2022:
1,689 models for proteins of human proteome are generated by D-I-TASSER-AF2 modeling. The distance prediction
is created by AlphaFold2, AttentionPotential and DeepPotential, the initial conformations are from AlphaFold2 models and LOMETS3 templates,
then a Replica-Exchange Montre Carlo simulation is used for constructing models. Multi-domain protein is first spilt by FUpred to domain-level
sequences, individual domain models are predicted by D-I-TASSER-AF2 modeling pipeline, then the first D-I-TASSER-AF2 domain-level
models are assembled by DeepDEMO to a full-length model.
Sep 1, 2021:
19,512 models for proteins of human proteome are generated by full D-I-TASSER modeling. The distance prediction
is created by AttentionPotential and DeepPotential, and initial conformations are from LOMETS3 templates, then a Replica-Exchange
Montre Carlo simulation is used for constructing 5 models. Multi-domain protein is first spilt by FUpred to domain-level
sequences, individual domain models are predicted by D-I-TASSER modeling pipeline, then the first D-I-TASSER domain-level
models are assembled by DeepDEMO to a full-length model.
May 1, 2021:
19,326 models for proteins of human proteome are generated by D-I-TASSER ab initio modeling. The distance prediction is created by DeepPotential, then a L-BFGS system is used for constructing 10 initial models. Those initial models
are then refined by D-I-TASSER to generate 5 models. Multi-domain protein is first spilt by FUpred to domain-level
sequences, individual domain models are predicted by D-I-TASSER ab initial pipeline, then be assembled by DEMO to a
full-length model.
Wei Zheng, Yang Li, Qiqige Wuyun, Xiaogen Zhou, Chengxin Zhang, Robin Pearce, Eric W. Bell, Yiheng Zhu, Yang Zhang. Integrating deep neural network models with I-TASSER for accurate protein structure prediction. Submitted. (2021). [PDF] [Support Information] [Database]
yangzhanglab
umich.edu
| (734) 647-1549 | 100 Washtenaw Avenue, Ann Arbor, MI 48109-2218