Improved Disorder Prediction by Combination of Orthogonal Approaches

By Avner Schlessinger, Marco Punta, Guy Yachdav, László Kaján, Burkhard Rost. 2009

190 citations recorded by the source corpus.

Disordered proteins are highly abundant in regulatory processes such as transcription and cell-signaling. Different methods have been developed to predict protein disorder often focusing on different types of disordered regions. Here, we present MD, a novel META-Disorder prediction method that molds various sources of information predominantly obtained from orthogonal prediction methods, to significantly improve in performance over its constituents. In sustained cross-validation, MD not only outperforms its origins, but it also compares favorably to other state-of-the-art prediction methods in a variety of tests that we applied. Availability: http://www.rostlab.org/services/md/

Subfields: Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

Citations in the corpus

Better prediction of functional effects for sequence variants (2015, neutral): “Additionally, we introduced new features for SNAP2: amino acid properties as provided by the AAindex database [42], predicted binding residues [43], predicted disordered regions [44], proximity to N- and C-terminus, statistical contact potentials [45], co-evolving positions, residue annotations from”
ESpritz: accurate and fast prediction of protein disorder (2012, background): “predictors, as adding more prediction layers has produced slightly et al. , 2010; Schlessinger et al. , 2009; Walsh et al. , 2011; Xue et al. , 2010).”
A community computational challenge to predict the activity of pairs of compounds (2014, unknown): “The first is a threshold-free metric designed to assess a method’s tradeoff between sensitivity and specificity in predicting synergistic or antagonistic combinations, whereas the second tests how precise in predicting the intended pairs the methods are at a specific cutoff.”
MFDp2 (2013, background): “39,42 The results are summarized in Table 1 .”
Comprehensive review of methods for prediction of intrinsic disorder and its molecular functions (2017, methodology): “The examples of the latter classifier-based consensuses are MD [74], MFDp [64], and MFDp2 [66] that use neural networks (MD) and support vector machines (both versions of MFDp).”
Computational prediction of secondary and supersecondary structures. (2013, background): “These characteristics include multiple alignment (17), prediction of protein-ligand interactions (18-20), prediction of residue depth (21, 22), structural classes and folds (23-25), residue contacts (26, 27), disorder (28-30), folding rates and types (31-33), and target selection for structural geno”
Improving protein disorder prediction by deep bidirectional long short‐term memory recurrent neural networks (2016, unknown): “Examples are CSpritz (Walsh et al. , 2011), MD (Schlessinger et al. , 2009), metaPrDOS (Ishida and Kinoshita, 2008), MFDp2 (Mizianty et al. , 2013) and PONDR-FIT (Xue et al. , 2010).”
Comprehensive comparative assessment of in-silico predictors of disordered regions. (2012, methodology): “Our assessment considers sixteen disorder predictors that cover the fours types of the methods, including four relative propensity-based methods including GlobPlot [[20]], IUPred [[21]], FoldIndex [[22]], and Ucon [[23]]; eight machine learning-based solutions such as DisEMBL [[25]], DISOPRED2 [[26]”
Intrinsically Semi-disordered State and Its Role in Induced Folding and Protein Aggregation (2013, unknown): “, MD [69] and MFDp [70].”
Improved sequence-based prediction of disordered regions with multilayer fusion of multiple information sources (2010, background): “Their prevalence was also implicated in various human disorders including the neurodegenerative diseases such as Huntington, Parkinson’s and Alzheimer’s disease (Raychaudhuri et al., 2009).”