%0 Journal Article %A Bohdan B. Khomtchouk %A Vytas Dargis-Robinson %A James R. Hennessy %A Claes Wahlestedt %T MicroScope: real-time statistical analytics platform and visualization engine for gene expression heatmaps %D 2016 %R 10.1101/034694 %J bioRxiv %P 034694 %X Motivation Large gene expression heatmaps (≥ 200 genes) are difficult to read: while general expression level patterns may be observed via hierarchical clustering, the identities of specific gene names become entirely unreadable at large scale. More importantly, current state-of-the-art heatmaps are entirely static, require programming skills to produce, and do not include any statistical significance information indicating the identities of differentially expressed genes.Results We create a user-friendly web app to create heatmaps that are both dynamic and interactive. A few mouse clicks allows a user to generate a heatmap based on an input file and proceed to dynamically navigate to any sector of the heatmap via zooming in and out to a specific region, cluster, or even single gene. In addition to visual magnification, MicroScope also allows users to analytically perform real-time statistical analyses with simple button clicks. Furthermore, MicroScope allows users to hover the mouse pointer over any specific gene to show precise expression level details, as well as create row and column dendrograms with colored branches using simple one-click features. As such, MicroScope presents a significant advance in heatmap visualization technology over current standard protocols.Availability and implementation MicroScope is hosted online as an R Shiny web app based on the D3 JavaScript library: https://microscope.shinyapps.io/microscope. The methods are implemented in R, and are available as part of the MicroScope project at: https://github.com/Bohdan-Khomtchouk/Microscope.Contact: b.khomtchouk{at}med.miami.edu %U https://www.biorxiv.org/content/biorxiv/early/2016/01/05/034694.full.pdf