This Shiny app is supported by the B2B Consortium Grant (http://www.benchtobassinet.com) and hosted on http://b2b.hci.utah.edu/shiny/mouse_heart_differentiation/
Joseph A. Wamstad, Jeffrey M. Alexander, Rebecca M. Truty, Avanti Shrikumar, Fugen Li, Kirsten E. Eilertson, Huiming Ding, John N. Wylie, Alexander R. Pico, John A. Capra, Genevieve Erwin, Steven J. Kattman, Gordon M. Keller, Deepak Srivastava, Stuart S. Levine, Katherine S. Pollard, Alisha K. Holloway, Laurie A. Boyer, Benoit G. Bruneau
Cell. 2012 Sep 28;151(1):206-20. https://doi.org/10.1016/j.cell.2012.07.035
Data is publicly available at https://b2b.hci.utah.edu/gnomex/gnomexFlex.jsp?requestNumber=7R
Transcript abundances were quantified using kallisto (Bray et al. 2016) and genome build
GRCm38 release 98 (jul2019.archive.ensembl.org). Estimated counts for all transcripts
per gene were summed to give a gene-level abundance estimation (Soneson et al. 2015).
Summed estimated counts were rounded to the nearest integer in order to run DESeq2 (Love et al. 2014) using a negative binomial LRT test with model (counts ~ time_point). This model tests in an anova-like way whether one timepoint differs from any other timepoint across the series.
Gene panel plots can be displayed using normalized counts or rlog values from DESeq2 or transcripts per million from kallisto. The light grey dot represents the mean expression value for that gene at each timepoint and the dark grey dots represent the value for each replicate. The vertical line at each timepoint depicts the range of the data and a line has been drawn to connect the mean at each timepoint to show the expression profile.
The heatmap displays z-scores (computed using DESeq2 normalized counts) for the selected gene list using a red-blue color scale.