A machine learning approach for gene expression analysis and applications

摘要: High-throughput microarray technology is an important and revolutionary technique used in genomics systems biology to analyze the expression of thousands genes simultaneously. The popular use this has resulted enormous repositories data, for example, Gene Expression Omnibus (GEO), maintained by National Center Biotechnology Information (NCBI). However, effective approach optimally exploit these datasets support specific biological studies still lacking. Specifically, improved method required integrate data from multiple sources select only those that meet investigator's interest. In addition, full power determine relationships among selected interpret meanings behind relationships. To address requirements, we have developed a machine learning based includes: • An meta-analysis sources; exploits information regarding context interest provided biologists. A novel cluster analysis identify hidden patterns representing between under conditions motif finding discovers, not common transcription factor binding sites co-regulated genes, but also miRNA associated with conditions. learning-based framework web application run tasks on online.