Zhao Yingdong, Sung Myong-Hee, Simon Richard
National Cancer Institute, National Institutes of Health, Rockville, MD, USA.
Methods Mol Biol. 2007;409:217-25. doi: 10.1007/978-1-60327-118-9_15.
Identifying epitopes that elicit a major histocompatibility complex (MHC)-restricted T-cell response is critical for designing vaccines for infectious diseases and cancers. We have applied two artificial intelligence approaches to build models for predicting T-cell epitopes. We developed a support vector machine to predict T-cell epitopes for an MHC class I-restricted T-cell clone (TCC) using synthesized peptide data. For predicting T-cell epitopes for an MHC class II-restricted TCC, we built a shift model that integrated MHC-binding data and data from T-cell proliferation assay against a combinatorial library of peptide mixtures.
识别能引发主要组织相容性复合体（MHC）限制的T细胞反应的表位对于设计针对传染病和癌症的疫苗至关重要。我们应用了两种人工智能方法来构建预测T细胞表位的模型。我们开发了一种支持向量机，使用合成肽数据预测MHC I类限制的T细胞克隆（TCC）的T细胞表位。为了预测MHC II类限制的TCC的T细胞表位，我们构建了一个整合了MHC结合数据和针对肽混合物组合文库的T细胞增殖试验数据的转移模型。
