期刊文献+
共找到1篇文章
< 1 >
每页显示 20 50 100
Improving <i>in Silico</i>Prediction of Epitope Vaccine Candidates by Union and Intersection of Single Predictors
1
作者 Ivan Dimitrov Darren R. Flower Irini Doytchinova 《World Journal of Vaccines》 2011年第2期15-22,共8页
The in silico prediction of peptide binding affinities to MHC proteins is a very important first step in the process of epi-tope-based vaccine design and development. Five MHC class II binding prediction servers were ... The in silico prediction of peptide binding affinities to MHC proteins is a very important first step in the process of epi-tope-based vaccine design and development. Five MHC class II binding prediction servers were combined in different ways and the resulting performance of these combinations was evaluated using a test set, which consisted of 4540 known HLA-DRB1 binders. The five servers were: NetMHCIIpan, NetMHCII, ProPred, RANKPEP, and EpiTOP. The top 5% of the ranked predictions from each server were combined using union and intersection operators. The outputs were evaluated in terms of sensitivity and positive predictive value (PPV). The union operator showed high sensitivity (65-79%) and low PPVs (6-8%), while intersection outputs had low sensitivities (4-41%) yet significantly higher PPVs (14-31%). Thus there is a defining trade-off between sensitivity and PPV for each combination. The union of outputs from different servers brings more “noise” than “signal” to the resulting set of predicted binders. Conversely, selecting only commonly predicted binders increases the probability that an identified binder is a true binder. 展开更多
关键词 MHC Class II Binders T-CELL EPITOPES hla-drb1 alelles
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部