Alternative splicing is a major source of diversity in mammalian proteins.  Actual information is derived from aligning sequences in the public database, but it is not enough to analyze mechanisms about alternative splicing events.  Recent development of the microarray technology helps us to focus on expression levels of each axon in a gene.  Expression data has an advantage over sequence data, because it contains information about co-expressed genes that control alternative splicing events.   However, the current splicing variant analysis requires custom-made microarray platforms.  We developed a method to detect splicing variants from publicly available gene expression data for offering information about alternative splicing events with co-expressed genes.