Templateâ based protein structure prediction in CASP11 and retrospect of Iâ TASSER in the last decade

Abstract

We report the structure prediction results of a new composite pipeline for templateâ based modeling (TBM) in the 11th CASP experiment. Starting from multiple structure templates identified by LOMETS based metaâ threading programs, the QUARK ab initio folding program is extended to generate initial fullâ length models under strong constraints from template alignments. The final atomic models are then constructed by Iâ TASSER based fragment reassembly simulations, followed by the fragmentâ guided molecular dynamic simulation and the MQAPâ based model selection. It was found that the inclusion of QUARKâ TBM simulations as an intermediate modeling step could help improve the quality of the Iâ TASSER models for both Easy and Hard TBM targets. Overall, the average TMâ score of the first Iâ TASSER model is 12% higher than that of the best LOMETS templates, with the RMSD in the same threadingâ aligned regions reduced from 5.8 to 4.7 à . Nevertheless, there are nearly 18% of TBM domains with the templates deteriorated by the structure assembly pipeline, which may be attributed to the errors of secondary structure and domain orientation predictions that propagate through and degrade the procedures of template identification and final model selections. To examine the record of progress, we made a retrospective report of the Iâ TASSER pipeline in the last five CASP experiments (CASP7â 11). The data show no clear progress of the LOMETS threading programs over PSIâ BLAST; but obvious progress on structural improvement relative to threading templates was witnessed in recent CASP experiments, which is probably attributed to the integration of the extended ab initio folding simulation with the threading assembly pipeline and the introduction of atomicâ level structure refinements following the reduced modeling simulations. Proteins 2016; 84(Suppl 1):233â 246. © 2015 Wiley Periodicals, Inc.