摘要
Serial Analysis of Gene Expression (SAGE) is a powerful tool to analyze whole-genome expression profiles. SAGE data, characterized by large quantity and high dimensions, need reducing their dimensions and extract feature to improve the accuracy and efficiency when they are used for pattern recognition and clustering analysis. A Poisson Model-based Kernel (PMK) was proposed based on the Poisson distribution of the SAGE data. Kernel Principle Component Analysis (KPCA) with PMK was proposed and used in feature-extract analysis of mouse retinal SAGE data. The computa-tional results show that this algorithm can extract feature effectively and reduce dimensions of SAGE data.
Serial Analysis of Gene Expression （SAGE） is a powerful tool to analyze whole-genome expression profiles. SAGE data, characterized by large quantity and high dimensions, need reducing their dimensions and extract feature to improve the accuracy and efficiency when they are used for pattern recognition and clustering analysis. A Poisson Model-based Kernel （PMK） was proposed based on the Poisson distribution of the SAGE data. Kernel Principle Component Analysis （KPCA） with PMK was proposed and used in feature-extract analysis of mouse retinal SAGE data. The computa- tional results show that this algorithm can extract feature effectively and reduce dimensions of SAGE data.
基金
Supported by the National Natural Science Foundation of China (No. 50877004)
