ABCPRED

ABCpred is a linear B cell epitope prediction tool based on artificial neural network model.
It works with Artificial Neural Network(ANN) and partial recurrent Neural Network(RNN).
B-cell epitopes were obtained from B cell epitope database (BCIPEP), which contains 2479 continuous epitopes, including 654 immunodominant, 1617 immunogenic epitopes. The dataset covers a wide range of pathogenic group like virus, bacteria, protozoa and fungi. Removing all the identical epitopes and non-immunogenic peptides, final dataset consists of 700 B-cell epitopes and 700 non-epitopes or random peptides (equal length and same frequency generated from SWISS-PROT).


To use the ABCPred tool and get more information, please click here.


Here are datasets used in developing ABCpred algorithm:


ANN

Multi-Layer Perceptron(MLP), also called Artificial Neural Network(ANN) is a wide network of parallel interconnections composed of adaptive simple units. It can simulate the interaction of biological nervous system to real world objects. The ANN consists of an input layer; an output layer and several hidden layers. Each neuron in each layer is connected to all neurons in the previous layer, and the input is the output of the neuron in the previous layer.


RNN

Recurrent Neural Network is usually used when sequential data is related to each other. The RNN has a loop join on the hidden state and this loop constraint ensures that the order information is captured


Citation:

Saha, S and Raghava G.P.S. (2006) Prediction of Continuous B-cell Epitopes in an Antigen Using Recurrent Neural Network. Proteins,65(1),40-48 PMID: 16894596