IMTECH Home nHLAPred  
A neural network based MHC Class-I Binding Peptide Prediction Server
Home ComPred ANNPred References Help Matrices Team Contact



Table S1:The total number of MHC binders and non-binders for various MHC alleles. The data of MHC alleles shown in green color has been used for training of neural network as well as quantitative matrices generation whereas data of MHC alleles shown in brown has been used only for generation of quantitative matrices. The number of MHC non-binders shown in bracket for HLA-A*0201 has been used for training of artificial neural network.

Table S2:The list of quantitative matrices of ComPred part of server obtained from literature.These Alleles are included to provide comprehensive platform for MHC prediction.

Table S3:The analysis of old quantitative matrices (obtained from BIMAS server) of MHC allelles for which hybrid apporoach has been developed in this study.The hybrid appoarch is based on the combined use of Artificial Neural Network and quantitative matrices.

Table S4:The summary of results obtained after filtering ANN prediction through old Quantitative matrices obtained from BIMAS server.

Table S4:The performance of nHLAPred for HLA-A*0201 on the data set of 128 peptides (19 high affinity and 27 intermediate-affinity binders of PRAME protein.The analysis has been done by using the ComPred part of server through different cutoff scores.


[ComPred] [ANNPred] [Links] [Algorithms] [References] [Help] [matrices] [Team] [Contact]