microscopic images [1]

Majority of papers contains microscopic images (of all types: live, fixed, wide-field, confocal, super-resolution, etc.). Seeing data in an image is a powerful tool to convince your audience of the point you are making. If the readers cannot see it they will be less likely to believe it and quantifications will be seen as not trustworthy. This post is dedicated to representing microscopic images in figures, specifically fluorescence based microscopy. Honestly, it is not possible for me to cover all of it in a single post, so this one is dedicated to single channel image only and there will be follow up posts.

Many authors show (what I consider a big mistake, and hope to convince you that it is the case) their images in the color of the fluorophore they used. And in that manner: GFP is always shown in green, Cherry, Tomato and others in red and CFP, DAPI, etc. in blue. By doing so, authors often jeopardize readers' chance to properly see the signals.

To explain I decided to use my own images of neurons and mitochondria to ensure good quality of the data. Images were adjusted in the same way and they only differ in applied color.




You can see that coloring an image affects your ability to see the structures, even though if you were to look at histograms of those images they would look exactly the same! The difference comes from different luminosity of colors. Luminosity, also refereed to as lightness, is how light/dark color is perceived (how sensitive our eyes are to that color). To understand it I converted red, green, and blue into a grey scale:



As you can see blue is in fact a very dark color with low contrast to black background. Along the light spectrum luminosity follows the curve (adapted from seos website):



Green/yellow are the colors we perceive as the brightest, but blues and reds are quite dark. Notice, that there is a difference between night and day vision and this is due to a different sensitivity of our cones and rods. This is of course a fairly simplified explanation and if you want to read more about it there is plenty of great literature around to dig into. I highly recommend starting with Margaret Livingstone's Vision and Art. Biology of Seeing, and you can follow with more "advanced" literature.

Coming back to the figures themselves, I would always recommend presenting them in black and white, as this gives the highest contrast:



You can see much more detail in those images and so will you audience. For single channel images it is also advisable to invert your images and depict the signal as black and background as white:



The black is less overwhelming in this last one and therefore the signal comes out as more clear. 

Even though black and white are my favorite ways of presenting images, there is an alternative/compromise solution. If you are truly dedicated to your colors and are not willing to give them up you can use the following approach: overlay black and white image with a colored one.

Remember, though that this affects the brightness of the signal, so it can be used only if you want to show structures, not compare signal
intensities!