MaxSub: an automated measure for the assessment of protein structure prediction quality

• DOI: 10.1093/bioinformatics/16.9.776
• Publication Date: 2002-07-26T22:55:00Z
• Authors: Naomi Siew; Arne Elofsson; Leszek Rychlewski; Daniel Fischer
• Keywords: CASP; Automated method

Abstract Motivation: Evaluating the accuracy of predicted models is critical for assessing structure prediction methods. Because this problem not trivial, a large number different assessment measures have been proposed by various authors, and it has already become an active subfield research (Moult et al. (1999) Critical methods protein (CASP): round III. Proteins , Suppl32–6). The CASP (1997) proteins II. Suppl3, Dedicated Issue, Moult Suppl32–6) CAFASP (Fischer CAFASP-1: fully automated 209–217) experiments demonstrated that difficult to choose one single, ‘best’ method be used in evaluation. Consequently, CASP3 evaluation was carried out using extensive set especially developed numerical measures, coupled with human-expert intervention. As part our efforts towards higher level automation field, here we investigate suitability automated, simple, objective, quantitative reproducible can automatic upcoming CAFASP2 experiment. Such should (a) produce single quality model (b) perform similarly evaluations. Results: MaxSub new independently further builds extends some introduced at CASP3. aims identifying largest subset \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(C_{{\alpha}}\) \end{document}atoms superimpose ‘well’ over experimental structure, produces normalized score represents model. there exists no models, easy evaluate how good measure is. Even though exact comparison straightforward, use test-bed extracted from fold-recognition models. A rough qualitative performance vis-a-vis shows agreement more accurate better predicting groups. expected, differences were observed among medium poor Overall, top six groups ranked are also We conclude suitable Availability: available at: http://www.cs.bgu.ac.il/~dfischer/MaxSub/MaxSub.html Contact: {nomsiew,dfischer}@cs.bgu.ac.il; [e-mail]; [e-mail] Supplementary Information: Full tables http://www.cs.bgu.ac.il/\batchmode \({\sim}\) \end{document}dfischer/MaxSub/MaxSub.html web site: http://PredictionCenter.llnl.gov/casp3/CAFASP http://www.cs.bgu.ac.il/~dfischer/CAFASP2 To whom correspondence addressed.