PromAn: an integrated knowledge-based web server dedicated to promoter analysis

Abstract

PromAn is a modular web-based tool dedicated to promoter analysis that integrates distinct complementary databases, methods and programs. PromAn provides automatic analysis of a genomic region with minimal prior knowledge of the genomic sequence. Prediction programs and experimental databases are combined to locate the transcription start site (TSS) and the promoter region within a large genomic input sequence. Transcription factor binding sites (TFBSs) can be predicted using several public databases and user-defined motifs. Also, a phylogenetic footprinting strategy, combining multiple alignment of large genomic sequences and assignment of various scores reflecting the evolutionary selection pressure, allows for evaluation and ranking of TFBS predictions. PromAn results can be displayed in an interactive graphical user interface, PromAnGUI. It integrates all of this information to highlight active promoter regions, to identify among the huge number of TFBS predictions those which are the most likely to be potentially functional and to facilitate user refined analysis. Such an integrative approach is essential in the face of a growing number of tools dedicated to promoter analysis in order to propose hypotheses to direct further experimental validations. PromAn is publicly available at http://bips.u-strasbg.fr/PromAn.

Figure 1

Flowchart of the PromAn integrated strategy. A single reference sequence and a set of orthologous genomic sequences are required as input. First, dinucleotide distribution, TSS and promoter location as well as TFBS predictions characterize the reference sequence. Next, a phylogenetic footprinting approach is used to determine the evolutionary conservation profile and TFBS evaluation with statistical scores. This integrated strategy is used to validate the TSS location, highlighting potentially active promoter regions and potentially functional TFBSs through the PromAn graphical user interface (GUI).

Figure 2

Example of a PromAn output. The upper frame displays the mouse rhodopsin genomic region (extracted from ±10 kb with respect to the start codon—red boxes), surrounding dinucleotide (AT in green and GC in red) and conservation (blue) profiles. The orange rectangle highlights the RPPR that is described in Figure 3. The lower frames depict the DBTSS, Eponine, GenScan, user-defined motifs (Match_Pattern), TRANSFAC (Match_TRANSFAC_MinSumGood), JASPAR (Match_JASPAR_CORE) and retina dedicated (Match_BIBLIO_Retina) predictions. Each prediction is displayed as a colored [gradient from low (grey) to high (red) Mean Distance score] box where the outline indicates the strand (blue for minus and red for plus).

Figure 3

TFBS analysis of the RPPR responsible for photoreceptor specificity. User-defined motifs, TRANSFAC, JASPAR and retina specific predictions are depicted. They highlight the conservation of the Ret-1/PCE-1, BAT-1, NRE and Ret-4 elements among mammals.