GPMiner

GPMiner is a computational tool that allows users to identify regulatory features involved in gene transcription initiation. It does this by identifying transcriptional start sites (TSSs), regulatory features, and co-regulated transcription factors in a group of genes. Typically, other computational methods for predicting TSSs or transcription factor binding sites (TFBSs) lack annotations for some important regulatory features such as CpG islands, tandem repeats, the TATA box, CCAAT box, GC box, over-represented oligonucleotides, DNA stability, and GC content. To address this limitation, GPMiner incorporates well-known prediction tools with orthologous conserved features to construct a novel system using a support vector machine.

The proposed system allows users to input a group of gene names/symbols, as well as an input sequence, and outputs promoter regions, regulatory features, and co-regulated transcription factors. According to five-fold cross-validation results, the predictive sensitivity and specificity are both roughly 80%. The benefit of this system is that it enables the co-occurrence of TFBSs to be determined and supports the investigation of gene transcriptional regulation.

Additionally, GPMiner can analyze an input sequence for homogeneity of experimental mammalian promoter sequences and observe conserved regulatory features between homologous promoters through cross-species analysis. The regulatory features are visualized graphically to facilitate gene promoter observations.