Online Services

●I-TASSER ●QUARK ●LOMETS ●COACH ●COFACTOR ●MUSTER ●SEGMER ●FG-MD ●ModRefiner ●REMO ●SPRING ●COTH ●BSpred ●SVMSEQ ●ANGLOR ●BSP-SLIM ●SAXSTER ●ThreaDom ●EvoDesign ●GPCR-I-TASSER ●GPCR-AIM ●BindProf ●ResQ

●TM-score ●TM-align ●MMalign ●NWalign ●EDTSurf ●MVP ●MVP-Fit ●SPICKER ●HAAD ●PSSpred ●3DRobot

●BioLiP ●E. coli ●GLASS ●GPCR-HGmod ●GPCR-RD ●GPCR-EXP ●TM-fold ●DECOYS ●POTENTIAL ●RW ●HPSF ●CASP7 ●CASP8 ●CASP9 ●CASP10 ●CASP11

[GPCR-AIM Server] [Forum]

About GPCR-AIM server

Comparative folding G protein-coupled receptors is obstructed by distant homology and limited experimental templates. To explore the possibility of ab initio modeling, we established an entirely ab initio approach, GPCR-AIM, which does not use any homology templates or fragments and employs cost-effective 2D and 3D helical topology simulating stages. The a new ab initio approach GPCR-AIM algorithm to assemble GPCR structure from the primary sequence. In GPCR-I-TASSER, although it also includes a component for assembling GPCR helix bundle from scratch, the procedure was guided by an over-simplified generic statistical potential, which is unable to accommodate the complicated architecture of helix arrangements. In GPCR-AIM, we use sequence-based contact map and disulfide-bond prediction, which are combined with a variety of GPCR-specific knowledge-based force fields to guide the helix-bundle assembly procedure. The entire ab initio modeling consists of four stages of 2D-helix optimization, 3D-helix bundle assemble, and reduced- and atomic structure refinement (see Figure 1 below), which has been shown to significantly improve the ability of the ab initio GPCR modeling of previous approaches.

The approach was evaluated in 28 solved GPCRs where GPCR-AIM correctly folded 21 (75%) targets (TM-score>0.5) and demonstrated more robust than homology approach and fragment-based ab initio approach on the families without enough experimental structures. Then, GPCR-AIM was applied to model 94 hard and unsolved GPCRs in the human genome, whose C-score<-1.5 and without any experimental data in GPCR-RD. Three types of UniProt annotations were used to investigate the quality of the GPCR-AIM models. 18 (31.0%) out of 58 UniProt annotated disulfide bonds are successfully detected, 56 (77.8%), 62 (86.1%) and 40 (58.3%) targets are correctly predicted for biological process (BP) function, cellular component (CC) function, molecular function (MF) function, and 65 (78.3%) targets are assigned to correct superfamily.

Figure 1. GPCR-AIM flowchart. (A) Four stages of GPCR structure assembly. (B) In Stage-2 helices are projected into the membrane plane, which are then reassembled by a weighted contact map potential. (C) Helix movements in Stage-3, where uninvolved helices are in gray and old and new conformations are in blue and yellow respectively. (D) Definition of helix-helix connection distances and cross-angles.

---

Please cite the following articles when you use the GPCR-AIM server:

• Hongjie Wu, Chengxin Zhang, Yang Zhang.GPCR-AIM: ab initio folding G protein-coupled receptors from 2D to 3D and application on hard GPCR human genome(preparing,2016)