A composite score for predicting errors in protein structure models Export

@article{citeulike:771073, abstract = {Reliable prediction of model accuracy is an important unsolved problem in protein structure modeling. To address this problem, we studied 24 individual assessment scores, including physics-based energy functions, statistical potentials, and machine learning-based scoring functions. Individual scores were also used to construct sim85,000 composite scoring functions using support vector machine (SVM) regression. The scores were tested for their abilities to identify the most native-like models from a set of 6000 comparative models of 20 representative protein structures. Each of the 20 targets was modeled using a template of \<30\% sequence identity, corresponding to challenging comparative modeling cases. The best SVM score outperformed all individual scores by decreasing the average RMSD difference between the model identified as the best of the set and the model with the lowest RMSD (DeltaRMSD) from 0.63 \r{A} to 0.45 \r{A}, while having a higher Pearson correlation coefficient to RMSD (r = 0.87) than any other tested score. The most accurate score is based on a combination of the DOPE non-hydrogen atom statistical potential; surface, contact, and combined statistical potentials from MODPIPE; and two PSIPRED/DSSP scores. It was implemented in the SVMod program, which can now be applied to select the final model in various modeling problems, including fold assignment, target-template alignment, and loop modeling.}, address = {Graduate Group in Biophysics, University of California at San Francisco, San Francisco, California 94158, USA; Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, and California Institute for Quantitative Biomedical Research, University of California at San Francisco, San Francisco, California 94158, USA; Departamento de Gen\'{e}tica Molecular y Microbiolog\'{\i}a, Facultad de Ciencias Biol\'{o}gicas, Pontificia Universidad Cat\'{o}lica de Chile, Alameda 340, Santiago, Chile}, author = {Eramian, David and Shen, Min-yi and Devos, Damien and Melo, Francisco and Sali, Andrej and Marti-Renom, Marc A.}, citeulike-article-id = {771073}, citeulike-linkout-0 = {http://dx.doi.org/10.1110/ps.062095806}, citeulike-linkout-1 = {http://www.proteinscience.org/cgi/content/abstract/15/7/1653}, citeulike-linkout-2 = {http://view.ncbi.nlm.nih.gov/pubmed/16751606}, citeulike-linkout-3 = {http://www.hubmed.org/display.cgi?uids=16751606}, citeulike-linkout-4 = {http://www3.interscience.wiley.com/cgi-bin/abstract/121603194/ABSTRACT}, doi = {10.1110/ps.062095806}, issn = {1469-896X}, journal = {Protein Science}, keywords = {assesment, model, modeller, score, structure}, month = {July}, number = {7}, pages = {1653--1666}, posted-at = {2008-01-29 20:04:38}, priority = {2}, title = {A composite score for predicting errors in protein structure models}, url = {http://dx.doi.org/10.1110/ps.062095806}, volume = {15}, year = {2006} }

TY - JOUR ID - citeulike:771073 L3 - citeulike-article-id:771073 N2 - Reliable prediction of model accuracy is an important unsolved problem in protein structure modeling. To address this problem, we studied 24 individual assessment scores, including physics-based energy functions, statistical potentials, and machine learning-based scoring functions. Individual scores were also used to construct sim85,000 composite scoring functions using support vector machine (SVM) regression. The scores were tested for their abilities to identify the most native-like models from a set of 6000 comparative models of 20 representative protein structures. Each of the 20 targets was modeled using a template of <30% sequence identity, corresponding to challenging comparative modeling cases. The best SVM score outperformed all individual scores by decreasing the average RMSD difference between the model identified as the best of the set and the model with the lowest RMSD (DeltaRMSD) from 0.63 Å to 0.45 Å, while having a higher Pearson correlation coefficient to RMSD (r = 0.87) than any other tested score. The most accurate score is based on a combination of the DOPE non-hydrogen atom statistical potential; surface, contact, and combined statistical potentials from MODPIPE; and two PSIPRED/DSSP scores. It was implemented in the SVMod program, which can now be applied to select the final model in various modeling problems, including fold assignment, target-template alignment, and loop modeling. IS - 7 JF - Protein Science SN - 1469-896X AD - Graduate Group in Biophysics, University of California at San Francisco, San Francisco, California 94158, USA; Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, and California Institute for Quantitative Biomedical Research, University of California at San Francisco, San Francisco, California 94158, USA; Departamento de Genética Molecular y Microbiología, Facultad de Ciencias Biológicas, Pontificia Universidad Católica de Chile, Alameda 340, Santiago, Chile EP - 1666 TI - A composite score for predicting errors in protein structure models VL - 15 SP - 1653 KW - assesment KW - model KW - modeller KW - score KW - structure AU - Eramian, David AU - Shen, Min-yi AU - Devos, Damien AU - Melo, Francisco AU - Sali, Andrej AU - Marti-Renom, Marc PY - 2006/07// UR - http://dx.doi.org/10.1110/ps.062095806 ER -