1 - 21 of 21 results


Predicts the relative surface accessibility of an amino acid and simultaneously predicts the reliability for each prediction, in the form of a Z-score. NetSurfP is composed of two neural network ensembles: the primary predicts secondary structure and have two outputs corresponding to buried or exposed; the secondary predicts the relative surface exposure of the individual amino acid residues. The tool is able to assign a reliability score to each surface accessibility prediction as an inherent part of the training process.

PredRSA / Prediction of Relative Solvent Accessible surface area

Predicts relative solvent accessible surface area (RSA) of residues. PredRSA integrates gradient boosted regression trees (GBRT) algorithm with multiple sequence-based features and a global side-chain environment feature. It uses a variety of multiple sequence-derived features, including the position specific scoring matrices and conservation score in the form of PSI-BLAST profiles, predicted secondary structure and natively disordered region.

PCASA / Protein-Ca Solvent Accessibilities

Predicts residue-wise Solvent Accessible Area (SASAs) from Ca coordinates of protein structures. PCASA is method that finds utility in accounting for solvation effects in coarse-grained protein folding simulations via SASA-based estimation of the transfer free energies. It was parameterized with a Bayesian linear regression model. It can also accurately predict residue-wise and total SASAs for conformations in an independent testing set. Importantly, PCASA was also shown not to be biased toward either folded or unfolded conformations.

MixMD / Mixed-solvent Molecular Dynamics

Accounts for the interaction of both water and small molecule probes with a protein’s surface. MixMD recognizes conserved and displaceable water sites. It employs an occupancy-based metric to find sites which are consistently occupied by water even in the presence of probe molecules. This method is able to determine which functional groups are capable of displacing which water sites. It stores the displacement of water sites by common functional groups.


Allows users to calculate the solvation free energy to arbitrary solvents, including inhomogeneous systems, and can run in cooperation with state-of-art molecular simulation softwares, such as NAMD, GROMACS and/or AMBER. ERmod (Energy Representation Module) is a software to calculate the solvation free energy based on the energy representation method. Molecular simulation is to be conducted at two condensed-phase systems of the solution of interest and the reference solvent with test-particle insertion of the solute. The subprogram ermod in ERmod then provides a set of energy distribution functions from the simulation trajectories, andanother subprogram slvfe determines the solvation free energy from the distribution functions through an approximate functional.


A method for predicting solvent accessibility and contact number simultaneously, which is based on a shared weight multitask learning framework under the CNF (conditional neural fields) model. The multitask learning framework on a collection of related tasks provides more accurate prediction than the framework trained only on a single task. The CNF method not only models the complex relationship between the input features and the predicted labels, but also exploits the interdependency among adjacent labels.


Predicts the real values of solvent accessibility in a protein using a neural network algorithm. In RVP-NET, single residue information of neighbours is used for making prediction of solvent accessibility. This method provides a direct prediction of ASA (accessible surface area) without making exposure categories and achieves results better than 19% mean absolute error. It provides a useful tool for the prediction of solvent accessibility and hence will be helpful in estimating structure and function of proteins with unknown three-dimensional structures.


Predicts solvent accessibility of amino acids using an optimized neural network algorithm. NETASA provides accuracy values, which are comparable or better than other methods of ASA (accessible surface area) prediction. Prediction in two and three state classification systems with several thresholds are provided. This prediction method achieved the accuracy level up to 90% for training and 88% for test data sets. NETASA also includes a linear activation function and some changes in the training procedure.

Sann / Solvent Accessibility prediction by Nearest Neighbor method

Predicts the discrete state (two and three states) as well as continuous value of the solvent accessibility of a target residue. The method is based on a k-nearest neighbor method combined with z-value distance statistics in the feature vector space. The overall performance of SANN is excellent and comparable to the best ones currently available. An independent benchmark test was performed with the CASP8 targets where we find that the proposed method outperforms existing methods. The prediction accuracies are 80.89% (for two state prediction with the threshold of 25%), 67.58% (three-state prediction), and the Pearson correlation coefficient of 0.727 (for continuous prediction) with mean absolute error of 0.148.

WESA / Weighted Ensemble Solvent Accessibility predictor

Predicts solvent accessibility of residues from protein sequences. WESA is based on five classification methods: the Bayesian statistics (BS), the multiple linear regression (MLR), the decision tree (DT), the neural network (NN) and the support vector machine (SVM) methods. It can serve to determine sites of protein hydration, which can play a role in a protein’s function. This tool can be useful to find sites of deleterious mutations.


Offers a platform for determining protein structural features and tertiary structures. SCRATCH is a web application including ten modules for determining three and eight class: (1) secondary structure, (2) relative solvent accessibility, (3) domain boundaries, (4) disordered regions, (5) disulfide bridges, (6) the effect of single amino acid mutation on stability, (7) residue-residue contact maps, and (8) tertiary structures as well as contacts with other residues compared to average.


Calculates the atomic accessible surface defined by rolling a probe of given size around a van der Waals surface. Naccess is able to calculate the atomic and residue accessibilities for both proteins and nucleic acids. The tool can be used for up to 20000 atoms, and allows the variation of the probe size and atomic radii by the user. It works by taking thin Z-slices through the molecule and calculating the exposed arc lengths for each atom in each slice, and then summing the arc lengths to the final area over all z-values.