List of disorder prediction software


Computational methods exploit the sequence signatures of disorder to predict whether a protein is disordered, given its amino acid sequence. The table below, which was originally adapted from and has been recently updated, shows the main features of software for disorder prediction. Note that different software use different definitions of disorder.
PredictorYear PublishedWhat is predictedBased onGenerates and uses multiple sequence alignment?Free for commercial use
2020Per-residue probability of a sequence residue being disordered.Ensemble of Bidirectional Long Short-Term Memory and Inception-Residual Squeeze-and-Excitation Convolutional Neural NetworksYesNo
2019Secondary structure and disorder prediction methodLong Short-Term Memory and Convolutional Neural NetworksYesNo
2018Per-residue disorder predictor for a single-sequence input.An ensemble of Long Short-Term Memory Bidirectional Recurrent Neural Networks and residual convolutional networks.NoNo
2005-2018Regions that lack a well-defined 3D-structure under native conditionsEnergy resulting from inter-residue interactions, estimated from local amino acid compositionNoNo
2017Consensus-based prediction of residue disorderEight separate disorder predictors from various groupsNoNo
2017Outputs the probability of each residue in a protein sequence of being disordered or ordered.A deep recurrent neural network architecture using Long Short-Term Memory cells.YesNo
2004-2015Regions devoid of ordered regular secondary structureCascaded support vector machine classifiers trained on PSI-BLAST profilesYesNo
2015Predict secondary structure and intrinsic disorder in one unified statistical framework based on the analysis of NMR chemical shiftsNeural networks trained on NMR solution-based data.YesNo
2015Assigns binary order/disorder class and corresponding confidence score for each protein residues using optimized SVM with Radial basis kernel from protein sequenceAA composition, Physical Properties, Helix, strand and coil probability, Accessible surface area, torsion angle fluctuation, monogram, bigram.No?
2014A binary prediction of whether a protein has a long disordered region Physicochemical properties of amino acids, sequence complexity, and amino acid compositionNo?
2013Helix, strand and coil probability, relative entropy and per residue disorder prediction.A combination of MFDp and DisCon predictors with unique post processing. Improved prediction over MFDp.YesNo
2012Disorder definitions include: missing x-ray atoms, Disprot style disorder, and NMR flexibility. A probability of disorder is supplied with two decision thresholds which depend on a user preferred false positive rate.Bi-directional neural networks with diverse and high quality data derived from the Protein Data Bank and DisProt. Compares extremely well with other CASP 9 servers. The method was designed to be very fast.NoNo
2012Regions that lack a well-defined 3D structure under native conditions Meta method, which uses other disorder predictors. Based on them the consensus is calculated according method accuracy. Currently the best available method YesNo
2012Output long/short disorder and semi-disorder and full disorder. Semi-disorder is semi-collapsed with some secondary structure.A neural network based three-state predictor based on both local and global features. Ranked in Top 5 based on AUC in CASP 9.YesNo
2011Disorder definitions include: missing x-ray atoms and DisProt style disorder. A probability of disorder is supplied with two decision thresholds which depend on the false positive rate. Linear motifs within a disorder segment are determined by simple pattern matching from ELM.Support Vector Machine and Bi-directional neural networks with high quality and diverse data derived from the Protein Data Bank and Disprot. Structural information is also supplied in the form of homologous templates. Compares extremely well with other CASP 9 servers.YesNo
1999-2010All regions that are not rigid including random coils, partially unstructured regions, and molten globulesLocal aa composition, flexibility, hydropathy, etc.NoNo
2010Different types of disorder including random coils, unstructured regions, molten globules, and REMARK-465-based regions.An ensemble of 3 SVMs specialized for the prediction of short, long and generic disordered regions, which combines three complementary disorder predictors, sequence, sequence profiles, predicted secondary structure, solvent accessibility, backbone dihedral torsion angles, residue flexibility and B-factors. MFDp YesNo
2005Regions that have a low hydrophobicity and high net charge Charge/hydrophaty analyzed locally using a sliding windowNo?
2005Regions that lack a well-defined 3D structure under native conditionsBio-basis function neural network trained on disordered proteinsNoNo
2003Regions with high propensity for globularity on the Russell/Linding scale Russell/Linding scale of disorderNoYes
2003LOOPS ; HOT LOOPS ; REMARK465 Neural networks trained on X-ray structure dataNoYes
1994Low-complexity segments that is, “simple sequences” or “compositionally biased regions”.Locally optimized low-complexity segments are produced at defined levels of stringency and then refined according to the equations of Wootton and FederhenNo?

Methods not available anymore:
PredictorWhat is predictedBased onGenerates and uses multiple sequence alignment?
The transition between structurally ordered and mobile or disordered amino acids intervals under native conditions.OnD-CRF applies Conditional Random Fields, CRFs, which rely on features generated from the amino acid sequence and from secondary structure prediction.No
Regions with No Ordered Regular Secondary Structure. Most, but not all, are highly flexible.Secondary structure and solvent accessibilityYes
Hydrophobic clusters, which tend to form secondary structure elementsHelical visualization of amino acid sequenceNo
Regions that are expected to be unstructured in all conditions, regardless of the presence of a binding partnerCompositional bias and low hydrophobic cluster content.No
Regions of different "types"; for example, unstructured loops and regions containing few stable intra-chain contactsA neural-network based meta-predictor that uses different sources of information predominantly obtained from orthogonal approachesYes
Long disordered regions in a set of proteinsMoreau-Broto auto-correlation function of amino acid indices No
Regions of different "types". MeDor provides a unified view of multiple disorder predictors.Meta method, which uses other disorder predictors and provides additional features that all together help the user in defining regions involved in disorder.No