I spend a lot of time making figures. To me, the ideal figure tells a clear story without compromising on nuance and transparency. In here, I intend to share resources that helped me make better figures, as well as best practices I follow. Like everything else in this site, this section is a work-in-progress.
I only make color blind friendly figures now. No more green and red in fluorescent images. And the good news is that you do not have to reinvent the wheel - the color palette I use is published in Nature Methods: https://doi.org/10.1038/nmeth.1618
I use ColorBrewer if I need colors beyond my usual palette: https://colorbrewer2.org
I had the privilege of TAing the class "Visualization for Persuasion" by Steve Franconeri at the Kellogg School of Management. In it, I learned evidence-based best practices for making data visualizations that are easier to follow and more memorable. Steve shares key takeaways from his research as a quick reference guide here: https://experception.net/
Variability comes in many flavors in science and it can hinder reproducibility. "SuperPlots" are a great way of transparently communicating sources of variability and efforts that were taken to make sure that the data is reproducible: https://doi.org/10.1083/jcb.202001064
I have not personally made a SuperPlot yet, but I always include individual data points in group data. The days of blindly comparing averages between two groups and displaying simple bar plots in figures are hopefully behind us!