 Thank you for joining me today. Do you see what I see? In this picture, I see an elephant, but not all people can. I'd like to introduce Micro Shades in our package for improving color accessibility in organization of complex data. Our motivation for this project was to create a color palette that's accessible to individuals that experience color vision deficiency, otherwise known as CVD or color blindness. CVD impacts roughly 300 million people worldwide, which is comparable to the current US population. CVD refers to partial but not complete color blindness and a diminished ability to distinguish between certain colors. There's three different types of CVD. Deuteranope is the most common and also known as red-green color blindness. Protanope is less common than deuteranope and described as a mutated red pigment with less ability to discriminate between colors. Tritanope is relatively rare and also known as blue-yellow color blindness. There are several CVD accessible color palettes available, one of which is the Kabaido color palette demonstrated here. As you can see, when run in different CVD simulations, this palette is universally CVD friendly. This is great, however, this palette and many others are only limited to eight colors, which is not sufficient for visualizing complex data. Our solution was to develop an R package that incorporates shading to expand on the number of colors that are CVD accessible. Our lab also works with complex microbiome data, so we wanted to make sure that we could use these colors with our data and created functions that would additionally aid in data organization. We developed two micro shade palettes, however, for the purpose of this talk, I will focus on the universally CVD palettes. This palette includes six base colors with different shading to result in a total of 30 colors. This slide here shows the micro shades CVD palettes under each type of CVD simulation. The colors are distinct and therefore universally accessible. Micro shades palettes can easily be accessed with their package and applied to any data plot. In this example, we're using the MPG data set available in R. This plot examines cars of fuel efficiency in terms of miles per gallon. The points on this plot represent a car with a particular type of drivetrain and cylinder engine. This first plot uses default coloring that's available in GG plot and provides no meaningful order to the coloring. This next plot uses the micro shades color in organization, each drivetrain is a different color and the number of cylinder engine is represented by a shade of that parent color. With this color organization, we're more easily able to see that front wheel drive vehicles tend to have the highest city and highway miles per gallon. Additionally, cars with fewer cylinders tend to have the most fuel efficiency. This color organization not only is accessible to all but additionally helpful in noticing these trends. Another example of micro shades in action is with this microbiome data example. This plot shows a relative abundance of different bacteria which are represented as colors in the samples which are represented as columns. This plot uses a default GG plot color scheme. This next plot shows the same data but with the accessible micro shade CBD pallets. The colors are organized and better represent the phylo genus classification of bacteria and a benefit can be seen when looking at this specific group of firm acutes that are circled here. You can see that the CBD colors improve the ability to distinguish these bacteria from each other in this plot. This last plot uses the micro shades accessible color pallets and also the organization functions. So the layout of the data further improves visibility of trends. Micro shades was also used to generate this custom legend that further shows the organization of the data. Thank you for joining my presentation. As you can see, there's many benefits of using micro shades. Please visit our website to learn how you can use micro shades with your own data.