Using Bayesian multinomial classifier to predict whether a given protein sequence is intrinsically disordered

Bulashevska, Alla; Eils, Roland

doi:10.1016/j.jtbi.2008.05.040

Cited by 11 publications

(9 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…IDP prediction after CASP7 exploited more methods, such as Spectral Graph Transducers (SGTs) [105], Bayesian methods [106], Conditional Random Fields (CRFs) [107], and metapredictors [61,[108][109][110].…”

Section: The First Period (1) First Informal Idp Predictormentioning

confidence: 99%

“…Bulashevska and Eils [106] were the first to apply this approach to predict IDPs. Each protein sequence belonging to a certain class can be considered as a realization of an independent random process that emits symbols from an alphabet of 20 amino acids.…”

Section: The First Period (1) First Informal Idp Predictormentioning

confidence: 99%

“…Ishita and Kinoshita developed the metaPdDOS metapredictor for per-residue estimates of order and disorder [106]. The MetaPdDOS uses a SVM to integrate residue-level predictions from PrDOS, DISOPRED2, DisEMBL, DISPROT, DISpro, IUPred, and POODLE-S.…”

Section: The First Period (1) First Informal Idp Predictormentioning

confidence: 99%

See 2 more Smart Citations

Predicting intrinsic disorder in proteins: an overview

et al. 2009

View full text Add to dashboard Cite

The discovery of intrinsically disordered proteins (IDP) (i.e., biologically active proteins that do not possess stable secondary and/or tertiary structures) came as an unexpected surprise, as the existence of such proteins is in contradiction to the traditional "sequence→structure→function" paradigm. Accurate prediction of a protein's predisposition to be intrinsically disordered is a necessary prerequisite for the further understanding of principles and mechanisms of protein folding and function, and is a key for the elaboration of a new structural and functional hierarchy of proteins. Therefore, prediction of IDPs has attracted the attention of many researchers, and a number of prediction tools have been developed. Predictions of disorder, in turn, are playing major roles in directing laboratory experiments that are leading to the discovery of ever more disordered proteins, and thereby leading to a positive feedback loop in the investigation of these proteins. In this review of algorithms for intrinsic disorder prediction, the basic concepts of various prediction methods for IDPs are summarized, the strengths and shortcomings of many of the methods are analyzed, and the difficulties and directions of future development of IDP prediction techniques are discussed.

show abstract

Section: The First Period (1) First Informal Idp Predictormentioning

confidence: 99%

Section: The First Period (1) First Informal Idp Predictormentioning

confidence: 99%

See 1 more Smart Citation

Predicting intrinsic disorder in proteins: an overview

et al. 2009

View full text Add to dashboard Cite

show abstract

“…DISOPRED [19] chooses 21 parameters per residue as input and metaPrDOS [20] takes meta approach integrating the results of seven different prediction methods to predicate disordered regions. Using Bayesian multinomial classifier, the predictive accuracy of 89.2% could be achieved for intrinsically disordered regions [21]. Although the current predictors have higher predictive accuracy [22], it is notable that there is a deficiency for the predictors on intrinsic disordered region.…”

Section: Sequence Characterizations Of Idpsmentioning

confidence: 99%

Novel Strategies for Drug Discovery Based on Intrinsically Disordered Proteins (IDPs)

Wang

Cao

Zhao

et al. 2011

IJMS

View full text Add to dashboard Cite

Intrinsically disordered proteins (IDPs) are proteins that usually do not adopt well-defined native structures when isolated in solution under physiological conditions. Numerous IDPs have close relationships with human diseases such as tumor, Parkinson disease, Alzheimer disease, diabetes, and so on. These disease-associated IDPs commonly play principal roles in the disease-associated protein-protein interaction networks. Most of them in the disease datasets have more interactants and hence the size of the disease-associated IDPs interaction network is simultaneously increased. For example, the tumor suppressor protein p53 is an intrinsically disordered protein and also a hub protein in the p53 interaction network; α-synuclein, an intrinsically disordered protein involved in Parkinson diseases, is also a hub of the protein network. The disease-associated IDPs may provide potential targets for drugs modulating protein-protein interaction networks. Therefore, novel strategies for drug discovery based on IDPs are in the ascendant. It is dependent on the features of IDPs to develop the novel strategies. It is found out that IDPs have unique structural features such as high flexibility and random coil-like conformations which enable them to participate in both the “one to many” and “many to one” interaction. Accordingly, in order to promote novel strategies for drug discovery, it is essential that more and more features of IDPs are revealed by experimental and computing methods.

show abstract

“…To date, a number of predictors of protein disordered regions have been published. From a view of algorithms which were used to construct the prediction model, several data mining and machine learning algorithms were applied, such as nearest neighbor algorithm [8], support vector machines (SVMs) [9–14], neural networks (NNs) [15–23], artificial neural network (ANNs), regression [24–26], sliding window [27,28], random forest [29], Bayesian Markov chain model [30] and so on.…”

Section: Introductionmentioning

confidence: 99%

A Novel Approach for Predicting Disordered Regions in A Protein Sequence

Cho

Ryu

2014

Osong Public Health and Research Perspectives

View full text Add to dashboard Cite

ObjectivesA number of published predictors are based on various algorithms and disordered protein sequence properties. Although many predictors have been published, the study of protein disordered region prediction is ongoing because different prediction methods can find different disordered regions in a protein sequence.MethodsTherefore we have used a new approach to find the more varying disordered regions for more efficient and accurate prediction of protein structures. In this study, we propose a novel approach called “emerging subsequence (ES) mining” without using the characteristics of the disordered protein. We first adapted the approach to generate emerging protein subsequences on public protein sequence data. Second, the disordered and ordered regions in a protein sequence were predicted by searching the generated emerging protein subsequence with a sliding window, which tends to overlap. Third, the scores of the overlapping regions were calculated based on support and growthrate values in both classes. Finally, the score of predicted regions in the target class were compared with the score of the source class, and the class having a higher score was selected.ResultsIn this experiment, disordered sequence data and ordered sequence data was extracted from DisProt 6.02 and PDB respectively and used as training data. The test data come from CASP 9 and CASP 10 where disordered and ordered regions are known.ConclusionComparing with several published predictors, the results of the experiment show higher accuracy rates than with other existing methods.

show abstract

Using Bayesian multinomial classifier to predict whether a given protein sequence is intrinsically disordered

Cited by 11 publications

References 31 publications

Predicting intrinsic disorder in proteins: an overview

Predicting intrinsic disorder in proteins: an overview

Novel Strategies for Drug Discovery Based on Intrinsically Disordered Proteins (IDPs)

A Novel Approach for Predicting Disordered Regions in A Protein Sequence

Contact Info

Product

Resources

About