June 10, 2021
by Charis Lam, Marketing Communications Specialist, ACD/Labs
Use ACD/Labs’ chromatographic optimization tools on your experimental data, and you’ll get a bright, vibrant, color-coded resolution map, like the following picture.
Bright orange shows the areas of highest resolution; deep blue shows the areas of lowest resolution; and the color scheme scales through green and yellow in between. At a glance, you can see where to focus your next experiments.
But what if the standard color scheme doesn’t work for you? Did you know that you can change it?
In this post, I’ll show some examples of when and how you might change your resolution map color scheme.
Reason 1: Viewing scientifically important distinctions (with how-to guide)
Reason 2: Making your resolution map more accessible
Reason 3: Printing in grayscale
Reason 1: Viewing scientifically important distinctions
The optimization tool’s default color scheme applies a gradual shift in color as areas go from low to high resolution. But you might consider some distinctions in resolution more important than others. For example, you might be satisfied with any resolution above 2, and having fine distinctions between light green at 2 and orange at 3 isn’t important. (You’d rather see finer gradations at lower resolutions.)
Imad Haidar Ahmad and colleagues at Merck made this point in a 2020 paper on multifactorial peak crossover.1 Resolutions greater than 1.5 were acceptable for their purpose. So they combined corresponding map areas (which were by default green, yellow, and orange) into one orange “good resolution” area. This gave better contrast with the blue dividing lines that marked crossovers.
Whatever your purpose, if you’re not interested in fine distinctions within some resolution range, you can also change your map to suit. Here’s how:
- Right-click on the resolution map to bring up the context menu. Select Resolution Map Coloring.
- The Resolution Map Coloring dialog box appears. To ensure the color scale matches the resolution scale, adjust the Min and Max fields at the bottom to cover the full resolution scale of the map. (Check the legend to the right of the map for the resolution scale.)
- The default color scheme is the Smooth Color Palette.
For this example, any resolution greater than ~3 is considered satisfactory. (Your exact numbers will differ based on your experimental needs.) So let’s compact blue, green, and yellow into a smaller area, so orange takes more space and represents all satisfactory resolution numbers.
- Beside the color palette is a set of RGB color sliders. Click on a few cells within the color palette and note their RGB values. For example, the rightmost cell on the fourth line from the bottom is teal and has an RGB value of (40, 168, 197).
Four rows up you have a green at RGB (112, 212, 34), and four rows up again you have a yellow-orange at RGB (255, 213, 0).
- You want to shrink this down, for example to take up 9 rows rather than 12.In the example, that would make the selected yellow-orange go from representing a resolution of 4.07 to representing a resolution of 3.05. (Again, your numbers may differ.)(Hover over a cell to see its represented resolution in the top left. This always corresponds to the cell you’re hovering over, not the cell you’ve clicked on.)
- Select the rightmost cell in the third row from the bottom. Use the RGB sliders to apply the first color you wrote down: RGB (40, 168, 197).Repeat on the sixth and ninth rows with the other two colors.
- Now you have three islands of color, but you want a smooth gradient. Select the cell at the bottom left (A, the deepest blue, representing the lowest resolution). Now select the first cell you recolored (B, the teal one). One cell will now have a dashed outline, and the other will have a solid one. These mark the endpoints of your gradient.
- Click the Gradient button (C). A smooth gradient appears from A to B.
- Repeat to add a gradient from B to the next changed color (three rows up), from that cell to three rows up again, and from that cell to the very last cell (top right).
- Make sure there’s a smooth gradient through the whole color scheme. Then click Apply. Your resolution map will update with your new color scheme.
- The color scheme will auto-save. Rename it by clicking Rename. It’ll be available whenever you open the Resolution Map Coloring dialog box, and you can apply it to a new figure with a few clicks.
Here’s a comparison of the default color scheme (left), the new one (middle), and an even more exaggerated one (with a lower threshold resolution). Now you can save and apply custom color schemes to any resolution map, to suit what you want to learn from your data.
Reason 2: Making your resolution map more accessible
You might also change color schemes to make data more viewable for your audience. For example, you might consider using an accessible color scheme. Color blindness affects about 1 in 12 men.2 (Prevalence is lower in women.) So if your graphic will be shown to a mid-size or larger audience, a color-blind person will likely be among them.
As red-green blindness is the most common form of color blindness,3 graphics that use red and green for important distinctions are most likely to be uninterpretable for some viewer. Accessibility advice4.5 suggests replacing rainbow palettes with red/blue or red/cyan.
The optimization tool’s default color scheme doesn’t use red and green together, but it can still be tweaked by simplifying the palette. For this example, I picked a diverging red/blue color scheme from ColorBrewer,5 an online color-advice tool for cartography.
I followed a similar procedure for setting the color scheme. Every few lines, I changed a cell to one of the colors in my new scheme. Then I applied gradients to smooth the entire palette out.
Diverging palettes like this work best if there’s a threshold value. In my example, a resolution of 2.01 is set to the lightest color (off-white), so there’s a clear distinction between values much less than 2 and much greater than 2. (You can change the threshold value to whatever applies for your project).
If divergence from a threshold value isn’t important, you could consider a monochromatic scale. Which brings us to Reason 3.
Reason 3: Printing in grayscale (or monochrome)
You might want a monochromatic scale for accessibility, or because you only have a black-and-white printer. (The process is the same, so we’ll group these together.)
Directly printing a multi-colored map won’t work. As the following picture shows, that completely flattens the distinction between green and orange areas. If that distinction is important, the graph is now far less helpful.
The optimization tool has a built-in grayscale color scheme. Open the Resolution Map Coloring dialog box again (see Reason 1 for how to find this). Select Gray Scale Palette. That gives the example on the left.
The contrast isn’t very high, so depending on how your map looks, you might want to increase it. Following the same steps as before, make the bottom-left cell of your color scheme very dark (near black) and the top-right cell white. Form a gradient between these two cells. That yields the example in the middle.
If you just want a monochromatic scheme, not necessarily grayscale, use a deep shade of any color instead of black. You can also reverse the scheme, so the deepest color represents highest resolution. (The legend will tell viewers which way the scale goes.) See the rightmost example.
Resolution maps in any color
By editing the color scheme, you can make resolution maps in any color. This isn’t just to make things look attractive, though attractive-looking graphics also have a purpose. (Audiences are attracted to well-designed graphics, and that applies to scientific figures too.)
Different color schemes might help you see important distinctions in your data. They might make your figures accessible to a wider audience. Or they might make them accessible to you, after you print them in black and white. The options are in your hands.
The resolution maps in this post were made using ACD/Labs’ chromatographic optimization tool, available in our Method Selection Suite and AutoChrom software.
1 Introducing Multifactorial Peak Crossover in Analytical and Preparative Chromatography via Computer-Assisted Modeling | Analytical Chemistry
2 Color Blindness | National Eye Institute
3 How to make scientific figures accessible to readers with color-blindness | American Society for Cell Biology
4 What to consider when visualizing data for colorblind readers | Datawrapper Blog