Prediction of structural features and application to outer membrane protein identification

Abstract

Protein three-dimensional (3D) structures provide insightful information in many fields of biology. One-dimensional properties derived from 3D structures such as secondary structure, residue solvent accessibility, residue depth and backbone torsion angles are helpful to protein function prediction, fold recognition and ab initio folding. Here, we predict various structural features with the assistance of neural network learning. Based on an independent test dataset, protein secondary structure prediction generates an overall Q3 accuracy of ~80%. Meanwhile, the prediction of relative solvent accessibility obtains the highest mean absolute error of 0.164, and prediction of residue depth achieves the lowest mean absolute error of 0.062. We further improve the outer membrane protein identification by including the predicted structural features in a scoring function using a simple profile-to-profile alignment. The results demonstrate that the accuracy of outer membrane protein identification can be improved by ~3% at a 1% false positive level when structural features are incorporated. Finally, our methods are available as two convenient and easy-to-use programs. One is PSSM-2-Features for predicting secondary structure, relative solvent accessibility, residue depth and backbone torsion angles, the other is PPA-OMP for identifying outer membrane proteins from proteomes.

DephosSite: a machine learning approach for discovering phosphotase-specific dephosphorylation sites

“…SVM is a popular machine learning algorithm and has been widely applied to solve many classification and regression problems in bioinformatics and computational biology 30 31 32 33 34 35 . Suppose we have a training vector x i ∈ R n and the class label y i = +1 or −1, i = 1, 2, …, l .…”

Identification of B‐ and T‐cell epitopes on HtrA protein of Orientia tsutsugamushi

“…15 The surface accessibility (buried/exposed) of the selected epitopes was calculated using BepiPred 2 program. The program utilizes both hydrogen bond energy and main chain dihedral angles (phi and psi), to derive secondary structures for structurally known proteins.…”

“…The α-helix, β-strand, coil, and turns were considered as described previously. 15 The surface accessibility (buried/exposed) of the selected epitopes was calculated using BepiPred 2 program.…”

“…The program utilizes both hydrogen bond energy and main chain dihedral angles (phi and psi), to derive secondary structures for structurally known proteins. The α‐helix, β‐strand, coil, and turns were considered as described previously . The surface accessibility (buried/exposed) of the selected epitopes was calculated using BepiPred 2 program.…”

Transmembrane region prediction by using sequence-derived features and machine learning methods

“…Here, two state-of-the-art machine learning algorithms (i.e., neural network and random forest) were used. As in our previous work, 30 the Encog 31 machine learning package was used to implement the neural network 32 algorithm. Briey, training was performed using the back-propagation algorithm as follows…”