MetaDisorder: a meta-server for the prediction of intrinsic disorder in proteins

BMC Bioinformatics

Research Article

Lukasz P Kozlowski¹  Janusz M Bujnicki² 

[1] Laboratory of Bioinformatics and Protein Engineering, International Institute of Molecular and Cell Biology, ul, Trojdena 4, 02-109, Warsaw, Poland;Laboratory of Bioinformatics and Protein Engineering, International Institute of Molecular and Cell Biology, ul, Trojdena 4, 02-109, Warsaw, Poland;Laboratory of Bioinformatics, Institute of Molecular Biology and Biotechnology, Faculty of Biology, ul Umultowska 89, 61-614, Poznan, Poland;

关键词: Protein Data Bank;    Area Under Curve;    Matthews Correlation Coefficient;    Primary Predictor;    Intrinsic Disorder;   

DOI  :  10.1186/1471-2105-13-111

 received in 2011-12-29, accepted in 2012-04-26,  发布年份 2012

来源: Springer

【 摘 要 】

BackgroundIntrinsically unstructured proteins (IUPs) lack a well-defined three-dimensional structure. Some of them may assume a locally stable structure under specific conditions, e.g. upon interaction with another molecule, while others function in a permanently unstructured state. The discovery of IUPs challenged the traditional protein structure paradigm, which stated that a specific well-defined structure defines the function of the protein. As of December 2011, approximately 60 methods for computational prediction of protein disorder from sequence have been made publicly available. They are based on different approaches, such as utilizing evolutionary information, energy functions, and various statistical and machine learning methods.ResultsGiven the diversity of existing intrinsic disorder prediction methods, we decided to test whether it is possible to combine them into a more accurate meta-prediction method. We developed a method based on arbitrarily chosen 13 disorder predictors, in which the final consensus was weighted by the accuracy of the methods. We have also developed a disorder predictor GSmetaDisorder3D that used no third-party disorder predictors, but alignments to known protein structures, reported by the protein fold-recognition methods, to infer the potentially structured and unstructured regions. Following the success of our disorder predictors in the CASP8 benchmark, we combined them into a meta-meta predictor called GSmetaDisorderMD, which was the top scoring method in the subsequent CASP9 benchmark.ConclusionsA series of disorder predictors described in this article is available as a MetaDisorder web server at http://iimcb.genesilico.pl/metadisorder/. Results are presented both in an easily interpretable, interactive mode and in a simple text format suitable for machine processing.

【 授权许可】

CC BY   
© Kozlowski and Bujnicki; licensee BioMed Central Ltd. 2012