Lens-PaTrNing : an interactive web resource for interrogating signal pathways and transcriptional networking during lens and cataract formation

Purpose: Defining the complex integration of signal pathways during tissue development and disease is a key frontier in systems biology. We have created a new web resource using public proteomic and transcriptomic data to predict novel molecular hypotheses of lens and cataract development. Methods: Public human and mouse lens gene expression datasets from various ages were combined with multi-species protein interaction data and cataract-associated genes using R. Commonly-expressed and lens-enriched genes were identified using compendium-based analyses. The data was combined into a web-based platform. Results: Lens-PaTrNing enables visualization of receptors, kinases and transcriptional regulators expressed by lens epithelial and fiber cells. This includes over 250 growth factor, cell-cell and cell-matrix signaling pathways. Correlation of these signaling pathways with target gene predictions allows extracellular-mediated control of lens gene expression to be investigated. The pathway/target gene data can be accessed: as receptor-defined pathways; by kinases and/or transcriptional regulators that are unique to individual pathways or common to multiple pathways; or by searching for individual genes to identify associated regulatory pathways. Each pathway contains interactive nodes that link to external resources. Side-by-side comparison of multiple pathways is possible, as is overlay of known cataract-associated genes on signal pathways and predicted target genes. This unique combination of interactive tools enables non-bioinformaticists to rapidly visualize lens signaling pathways and their involvement with known cataract-associated genes. Investigation of these signal pathway/target gene predictions identified over 100 pathways that regulate at least 50 cataract-associated genes, including critically-required lens pathways (e.g., Fgf), transcription factors (e.g., Pax6) and known cataract-associated genes (e.g., crystallins). Conclusions: The Lens-PaTrNing web interface is a powerful new public tool that enables non-bioinformaticians to visualize extrinsic and intrinsic lens signal pathways, their interconnections, and associated predicted target genes. In this way Lens-PaTrNing facilitates simple generation of candidate molecular mechanisms that can form the basis for hypothesis creation and hypothesis-driven investigation of lens and cataract formation.