Ab initio modeling of small proteins by iterative TASSER simulations

By Sitao Wu, Jeffrey Skolnick, Yang Zhang. 2007

437 citations recorded by the source corpus.

BackgroundPredicting 3-dimensional protein structures from amino-acid sequences is an important unsolved problem in computational structural biology. The problem becomes relatively easier if close homologous proteins have been solved, as high-resolution models can be built by aligning target sequences to the solved homologous structures. However, for sequences without similar folds in the Protein Data Bank (PDB) library, the models have to be predicted from scratch. Progress in the ab initio structure modeling is slow. The aim of this study was to extend the TASSER (threading/assembly/refinement) method for the ab initio modeling and examine systemically its ability to fold small single-domain proteins.ResultsWe developed I-TASSER by iteratively implementing the TASSER method, which is used in the folding test of three benchmarks of small proteins. First, data on 16 small proteins (< 90 residues) were used to generate I-TASSER models, which had an average Cα-root mean square deviation (RMSD) of 3.8Å, with 6 of them having a Cα-RMSD < 2.5Å. The overall result was comparable with the all-atomic ROSETTA simulation, but the central processing unit (CPU) time by I-TASSER was much shorter (150 CPU days vs. 5 CPU hours). Second, data on 20 small proteins (< 120 residues) were used. I-TASSER folded four of them with a Cα-RMSD < 2.5Å. The average Cα-RMSD of the I-TASSER models was 3.9Å, whereas it was 5.9Å using TOUCHSTONE-II software. Finally, 20 non-homologous small proteins (< 120 residues) were taken from the PDB library. An average Cα-RMSD of 3.9Å was obtained for the third benchmark, with seven cases having a Cα-RMSD < 2.5Å.ConclusionOur simulation results show that I-TASSER can consistently predict the correct folds and sometimes high-resolution models for small single-domain proteins. Compared with other ab initio modeling methods such as ROSETTA and TOUCHSTONE II, the average performance of I-TASSER is either much better or is similar within a lower computational time. These data, together with the significant performance of automated I-TASSER server (the Zhang-Server) in the 'free modeling' section of the recent Critical Assessment of Structure Prediction (CASP)7 experiment, demonstrate new progresses in automated ab initio model generation. The I-TASSER server is freely available for academic users http://zhang.bioinformatics.ku.edu/I-TASSER.

Subfields: Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

In Molecular Biology

Citations in the corpus

I-TASSER server: new development for protein structure and function predictions (2015, positive): “I-TASSER-based algorithms have been extensively tested in both benchmark (8,10,23,33) and blind tests (26,27,29,34).”
Template-based protein structure modeling using the RaptorX web server (2012, neutral): “Pure ab initio methods 17–19 aim at building a 3D model without using structure homologous information; the successful application of such methods is, however, limited to short target sequences ( < 120 residues) at present.”
A position-specific distance-dependent statistical potential for protein structure and functional study (2012, neutral): “…Shakhnovich, 2006) in many applications including ab initio folding (Jones and Thirup, 1986; Kihara et al., 2001; Levitt, 1992; Simons et al., 1997; Wu et al., 2007a; Zhao et al., 2008, 2010), docking (Zhang et al., 1997), binding (Kortemme and Baker, 2002; Laurie and Jackson, 2005), mutation…”
LOMETS: A local meta-threading-server for protein structure prediction (2007, neutral): “First, the I-TASSER force ﬁeld includes a variety of knowledge-based, protein-sequence speciﬁc/ nonspeciﬁc potentials obtained from variant resources (31), which has been optimized using structure decoys.”
Ab Initio Protein Structure Assembly Using Continuous Structure Fragments and Optimized Knowledge-based Force Field (2012, neutral): “5,14 The importance and difficulty of ab initio benchmarking were also reflected in the community-wide blind CASP experiments,19,20 where the portion of FM targets has been consistently decreasing owing to the difficulties in collecting individual protein domains which have different folds from the…”
De novo protein structure prediction by dynamic fragment assembly and conformational space annealing (2011, methodology): “(2) All trial conformations are prescreened to select only near-native structures.”
Protein Structure Prediction: Is It Useful? (2009, positive): “[35] developed a new version of I-TASSER that aims at refining the structures by an iterative implementation of the TASSER assembly procedure.”
Progress and challenges in protein structure prediction. (2008, positive): “[46 ], which refines the TASSER cluster centroids by iterative Monte Carlo simulations.”
I-TASSER: a unified platform for automated protein structure and function prediction (2010, neutral): “Finally, for query proteins that have no structurally related protein in the PDB library, the structure must be built from scratch by ab initio modelin”
Improving the physical realism and structural accuracy of protein models by a two-step atomic-level energy minimization. (2011, positive): “We test ModRefiner on two sets of proteins with the initial Ca trace models generated by I-TASSER (3,23), a typical algorithm based on multiple template assembly.”
Recognizing protein substructure similarity using segmental threading. (2010, methodology): “TASSER (Zhang and Skolnick, 2004a) and I-TASSER (Wu et al., 2007; Zhang, 2009) excise continuously aligned structural fragments ( 21 residues on average) from threading templates which are then reassembled by Monte Carlo simulations with the purpose of refining the template structures.”
Serum Albumin Domain Structures in Human Blood Serum by Mass Spectrometry and Computational Biology (2015, background): “Typically, searching the conformational space for the native structure is difficult because of the size of the space and the ruggedness of the energy landscape, although for small proteins this has been achieved (47).”
Template-based protein structure prediction in CASP11 and retrospect of I-TASSER in the last decade (2015, neutral): “I-TASSER9,20 was designed to construct protein structural models by reassembling continuous fragments excised from the top LOMETS threading alignments.”
MUSTER: Improving protein sequence profile–profile alignments by using multiple sources of structure information (2008, neutral): “In Table I, we present the results of three programs (PPA, MUSTER1, and MUSTER2) on the two training protein sets.”
Improving protein structure prediction using multiple sequence-based contact predictions. (2011, positive): “The contacts are then used as restraints in I-TASSER simulations (Wu et al., 2007).”
QMEAN server for protein model quality estimation (2009, neutral): “Such sets of models are for example obtained from meta-servers (16,17) (i.e. servers collecting and integrating results from multiple modelling servers) or as a result of extensive conformational sampling runs typically performed in fragment-based approaches (18,19).”
In silico methods and tools for drug discovery (2021, neutral): “When neither is available, in silico prediction methods such as homology modeling [16] or ab initio modeling [17] can be used to predict the 3D structure of the target protein.”
Recognizing protein-ligand binding sites by global structural alignment and local geometry refinement (2012, positive): “For a target protein, the structure models are first generated by the automated I-TASSER structural assembly method (Roy et al., 2010; Wu et al., 2007).”