Recently, biological databases and analytical methods have become available for analyzing gene expression and transcriptional regulatory sequences. However, users must make the complicated analyses to query the data in various databases, and then they must analyze the gene upstreams using various predictive tools, before finally converting date among formats. Beyond methods for predicting transcriptional regulatory sites, new automated and integrated methods for analyzing gene upstream sequences on a higher level are urgently required. Efficient and integrated data management methods are essential, too. We present an integrated system, namely RgS-Miner, to predict transcriptional regulatory sites and detect co-occurrence of these regulatory sites. RgS-Miner comprises a biological data warehousing system, pattern discovery programs, pattern occurrence association detectors and user interfaces. The system is available at http://rgsminer.csie.ncu.edu.tw/.