BioRender Top Grant Tips
Shiz Aoki, CEO & Co-founder of BioRender, will give her top 5 figure and design tips for creating grant proposals.
All right, excellent. So this will be a dive into a few tips I have, so much that I wanted to share, but I thought for the sake of time and focus, I wanted to make sure that we leave you with something that you can remember at least to make sure your Grant figures are beautiful next time, and hopefully we will overcome some of those cringe-worthy design flaws that were just discussed in the amazing panel just now. I am using BioRender, but of course these tips are pretty universal, so no matter what tool you're using, you'll be able to apply these tips. But you know, shameless plug, I am going to be using BioRender as well. We're using the BioRender slides feature. I know this is relatively new, not many of you may have known about it yet, but you'll see that I can actually use BioRender as a multi-figure composing canvas in a sense. So along the right-hand panel here, I'm going to be using BioRender to present a design within the slide feature. So it's a little bit meta, but bear with me here.
This will again be sort of a design-focused webinar. I would say that generally, as you already heard some of the speakers talk about, it's really about focusing on your Grant figure. So I believe that all the same design principles can sort of be applied to a messy room in a sense. Any lack of focus, of color, let's say there's too many things in the room, things in the wrong place, no sort of focal point. I'm sure you can get that feeling when you look at your figure or another person's figure that feels cluttered, almost conjuring up memories of a messy room. So we'll apply some of those principles as we go through these tips here. And of course, you can get experimental or adventurous with color, but you have to use color very sparingly. In this example, these very bold sort of turquoise color options, you know, they're usable, you can incorporate it, and in fact, we encourage it, but again, very sparingly and be selective so that those are the areas that are going to draw your reviewer's eye to that part of your figure. So I hope that makes sense. When you're using color, use it very, very sparingly.
I'm getting a question here that you don't see half the screen. Are we looking okay still? Let me know if it's not. It looks good to me, yeah? Okay, all right, thanks for the heads up anyway.
And as a heads up, I know Dr. Miranda did discuss this already, using the overall end product. So whether it's a specific aim or a certain layout that you have to follow, making the figure is all well done good, but it has to translate within the overall design like this. So that usually means that once you're finished your figure, you know, I personally would shrink it down into sort of dummy text like this so that you know that it'll register at whatever size the specs require it for. You're making a figure on your large screen or a large monitor, but eventually it actually has to squeeze down to the size of a business card. So there's no point in making it so detailed if you actually won't register those details in its final form. Just keep that in mind. I kind of screen grabbed Dr. Murin's image there, so I hope that's okay, but I thought it was a fantastic example of the basic principles of a very cohesive color palette. You notice that he used a different font here for the caption versus the font in the specific aims section here in the main body text. If you rewind the recording to Dr. Murin's talk, you'll see his explanation of his layout here, but I wanted to call that out specifically as a great example.
For the next say 20 minutes, I know time flies by when we're having fun, I'm going to focus on two tips. Two things that I want you to sort of take away from this is to improve the reading order of your diagrams or figures and also to be able to create focus in your figure with really a couple of basic sub-tips there. So your reading order and creating focus, and if you can do those two things to improve your figure for your Grant application, I think it'll really sing.
First thing is reading order left to right, up to down or clockwise. I think this feels pretty self-explanatory, but it's something that we commonly see as a mistake in a lot of figures. So looking at figures, they should always read either left to right, up to down, maybe cyclical if it's some sort of cyclical process.
Repeatable process: I would say that for specific aims, whether it's two, three, or one aim, you know as the argument was saying, it should still flow from left to right, up to down. You shouldn't get this sort of windy road to get the reviewer to sort of maze their way through your figure. It should be very simple.
By the way, I'm using Annotate as a plug-in here on Google Chrome. This is not part of the BioRender app; I'm just using it for presentation purposes. So that's the reading order. You know, maybe you select one or two of these maxims when you're putting together your figure. Here's an example. We had a few brave volunteers submit their grant figures to be reviewed live like this. It's sort of like live office hours.
What I like to do is start from the beginning to the end. If I were to track my pupils or retina to see where I'm looking when I approach a figure, usually it's the top left. And in this case, I'm seeing arrows which are going to sort of direct me to certain areas of my figure. Two arrows there. Now I'm seeing an arrow come back, so it's sending me back in this loop. And there's something going on down here. I think they want me to end here. So if you look at the path of least resistance, there's a pretty high friction here to get to the beginning and end of the story. And this is a really good exercise to do on any figure - just trace either your finger or maybe use a marker like this and see where your eye is led. This is actually directly correlated to probably the exhaustion that your reader will experience and trying to interpret your figure.
So recreated, we actually just rotated the figure about 90 degrees. There's always a way to better lay out your figure, so in this case, we took the figure, flipped it on its side, and repositioned a couple of the items here. So all the content is really the same - not a lot more to do, really. It's just about the layout and, again, the directionality of the reading order. So now we've got a nice waterfall happening top down. We do have some arrows pointing upward, but it's not as intense as previously where it was taking me all around the page. And generally speaking, it looks like this applicant or scientist wants me to read this figure from top down. That quick change made the reading order really clean and concise, keeps it focused, and again, doesn't tire out your reviewer because again, they're going to be reading stacks of grants, and they're not going to want to go back and try to interpret your figure if it's sort of taking you on a maze.
That's the reading order. This is another figure example. If your image that you're going to be presenting or submitting is a little bit more cellular like this, again, be careful where you're placing numbers and arrows. If I were to trace one, two, three, to four, you're seeing that my eye is getting looped around to all different parts of the figure. If we were to re-jig this a little bit, we flipped it over sort of 90 degrees, and this took a little bit more puzzle piecing, but in general, there's always a way to redesign content to be a little bit more smooth. So what we did was we put the story to start from the top left. That's usually a good place to start. At the top left and then have your story flow in almost a cyclical pattern. And what we did was we created larger arrows. We rejigged the story so that you've got one, two, three, and four generally following a circle pattern. It's not perfect, but it still has a bit of that cyclical flow that will allow your eye to more smoothly follow the story. And sometimes you'll have smaller sub-stories, and that's okay. But maybe differentiate that with smaller arrows. If I clear away my arrows here, you'll see that the biggest gravitational pull is around the edges, and then we've got these smaller mid-stories happening inside, which is again okay. But if the thickness and weight of your arrow is the same all throughout, then your reviewers won't really know what is the main focus of your story.
Okay, so that tip was covering the reading order. So left to right, up to down, or clockwise. So be sure to follow that. And then finally, to create focus in your figure, because a lot of the time, people will come to us and say, "My figure feels very cluttered, and I don't know what to remove or how to redesign it." People tell me to leave a void space, but I don't know how. So here are some tips that you can follow to create focus in your figure.
Color temperature and color saturation
Transparencies and being audience specific so this will mean something to you hopefully as we go through each.
Color temperature is pretty self-explanatory. What is temperature? It's hot and cold. That's basically what colors can be attributed to. So, if I were to take these objects, I'm just going to option drag in BioRender, I can actually duplicate icons doing that as a shortcut. There's a rule of thumb in the design world where cool colors tend to be more soothing to the eye. Sometimes protagonists of a story tend to be colored in blues, purples, or greens, especially in scientific figures. Warm colors tend to suggest antagonists of the story or sort of the bad guys or you know inflammation or blood things that denote negative impacts in the body.
So, what I would suggest is to flip the script, flip the story. If you have a cancer cell that's cool in color, we're actually going to create a strange story here where that ends up looking like the protagonist. So, it's pretty simple. What you want to do is make sure that all of your I guess lymphocytes in this case or a cooler color and then anything that's sort of the antagonist industry, make sure that those are colored to be a little bit warmer. So, you can see that right away, I'm getting a register of the lymphocytes being the good guys in the story and then the cancer cell being sort of the Invader whereas if these didn't have labels, you'd kind of automatically assume the opposite. So, really simple tip, make sure to use warm and cool colors to your advantage and then in some cases where there's an autoimmune response or something like that, you might want to color the lymphocytes a sort of warm color to show that this is actually not what you want happening in the body and you want to reverse that process.
All right, so that was the color temperature. Now, color saturation is also something that you can play with, turn the dial on to create focus in your figure. Here's an example of a very simplified sort of pathway diagram. In this case, it kind of reminds me of a room that has yellow accent pillows and a red curtain and maybe like a blue bookshelf. But basically, what happens is when you color everything a unique color, then nothing becomes important. So, what I recommend is really dial back, almost wipe the slate clean of color, and reintroduce color slowly.
So, here we go. Start from scratch. I've reduced the color. This is the exact same figure, by the way. Nothing has changed except the color. Isn't that crazy? Such a different registry of information. So, I've removed the color. I've incorporated color in just the top part. So, that's where my eye went right away. That's the highest saturation. Saturation is basically the brightest or boldest color versus unsaturated, which is going towards sort of grayscale or black and white. Slowly start to reintroduce color to areas that your focus is. Is it the TGF beta? Is your focus these proteins on the left? Maybe your focus is more within the DNA, or the nuclei rather. So, depending on where you're focusing your story, that's where I would focus the saturation and even contrast. But saturation really can be your best friend here if you use the tool properly.
And actually, with BioRender, you can very easily change the color of these sorts of text bubbles. If you are BioRender experts, you know that already, but we have pre-made color palettes that make it really easy to select really any color under the rainbow. And then the font will actually change to the opposite value. So, it'll be white on dark or black text on light background colors. So, play around with that really neat feature.
Then another trick you can use within BioRender is to preview in grayscale. Not everybody knows about this tip, but I use it very, very often. I'd say if you walk into any professional Design Studio, everyone will have grayscale versions of their design because it's really important to make sure it's accessible, colorblind accessible but also just generally accessible to even fully, you know, visionable folks because sometimes colors are deceiving.
So, in this case, you can clearly see the red cells on top of this tube, but these cells are getting lost. If I were to preview in grayscale, you can see that it very quickly tells me yes indeed it is disappearing. So, you might be thinking, okay, well, why don't I just make it a different color? So, you go ahead and do that, maybe make it an orange. Go back to grayscale, still it disappears, and that's because it's still too similar of a color value, basically the lightness or darkness is too similar to the background. So, your foreground elements have to be distinct enough of a darkness, not just a different color but a completely different lightness or darkness. So, maybe I want to pick a cell that is very bold blue or very bold red or orange. Basically, if you want to use orange, that's fine, but you have to make sure that it's a different color value. So, let's try that again. It's a little better, but you can see you can kind of toggle in black and white just to make sure that you have enough contrast in your figure. And this is going to make a world of a difference when your reviewer comes to your figure and is actually able to register the information in it. Here's another classic example of nuclei of cells that's staying dark. You know everything that you layer on top of those nuclei are going to actually disappear if it's very similar in color value. So, this is that trick again where we're going to preview in grayscale. I know in the Fine Art World, we tend to squint. You actually close one eye and squint to see if you can actually see the figure. Sometimes I'll even zoom out in my BioRender canvas like that. And if you can't see it at that scale, it probably means that your viewer is going to miss it as well. So, this is kind of a trick that I like to use. I'm actually rolling with my mouse wheel to zoom in and out. You can actually just use these to reset zoom or use the slider to zoom in and out, but this is a good trick to use. And actually, if your figure is going to be pretty small in its end-state, maybe a good idea to even zoom out just to see what it's going to register as.
So, in this case again, what do you think we could do here? Probably lighten the cell color, right, so that the foreground elements are more clear. That went from looking quite dark to nice and contrasty. Or this is a little bit trickier, but what you could do is make the foreground elements much lighter. Now, this is a little bit more tricky, and you could get in trouble doing this, but sometimes it does work. See how I've made the DNA much lighter there? Still having issues with these orange molecules, so I would probably go back in and again make those a stronger color against that background image.
Okay, that's previewing in grayscale to check for contrast issues. If there's one thing you take away from today's design tips here, it would be that. So, definitely keep that in mind. All right, the last few tips here are to look at transparencies and audience-specific content. So, I'm going to group those two together in the next examples.
I like to show this example because this shows a lot of information, but I suspect that your grant reviewers, as much as a specialist as you are in your science, there's going to be things that are maybe canonical or universal, like you don't really have to hit them over the head with, for example, you know these are indeed lungs. I think everyone can agree on that, and that's pretty well established. In this case, you can actually afford to reduce the transparency of them in favor of showing the trachea and the bronchioles a little bit more clear because again, you don't have to shout at them that these are lungs.
Same thing with the trachea actually. You can probably reduce the transparency or rather increase the transparency, reduce the opacity. You see how the lymph nodes become so much more apparent when I sort of slide the transparency up and down? And now right away the nodes sort of stand out to me much more than before because audience-specific, again, I don't have to teach them that there are trachea and lungs in the body. They're suggested there just for anatomical placement, but again, you don't need to shout at them. Same with the liver. I can't really see this metastasizing tumor because the liver is so dark. So, what I would do is don't be afraid to play with the transparency slider.
Another way to do this is to sort of have an outline of the organs, but I think in this case, this works pretty well. So that was really simple, right? Just kind of playing with the opacity of layering layered objects so that your main story really stands out. And again, the focus is now the tumor: three centimeters, two to five, five to seven, and over seven. I bet you that didn't really stand out before when the lungs were at full opacity. Okay, so that was another quick tip there. I'm going to quickly go through in the last five minutes a couple of more examples that were volunteered bravely, and we'll see how the design tips that we just learned apply to making them better.
I loved this proposal/figure because it was very nicely sort of aligned text-wise. There was some use of color coordination, so anything related to regulators of prmt5 activity were color-coded in blue, and same thing with green. One thing our team did to sort of make over this figure was to again dial back the amount of colors being used, and so this is the after. This is the before. Pretty subtle difference, but you can see that we didn't need to incorporate all those colors, same with the background. The nucleus didn't have to be purple. Thinking about audience specificity, they probably don't need to know that this is a phospholipid bilayer. I think generally speaking, most reviewers will know that that is a cell membrane, so we opted for this sort of smoother, double-lined iconography, and that automatically cleaned it up, cleaned up the composition a little bit. Not everything has to be in its own box all the time. I know we're tempted to box things in, but that can actually work against you, so we removed that box, and generally speaking, I think it flows a little bit better. Maybe I would sort of tweak and move things around just a little bit, but I think for the most part, the before and after looks quite nice. I don't know what you all think, but I think there's definitely an improvement there.
Here's another example, a little bit more of a cellular process, but thinking about audience specificity. You know, this is, yes indeed, a cell with an outline and cytosol, but right now, again, if I squint, if I do the old the squint trick, right now, I see pink, a bright orange membrane, and a yellow interior with some stuff happening and this little Greens on the side that's really standing out to me. Again, as a reviewer, I probably don't need to be shouted at that this is the edge of the cell. This could be really muted and have the inside, whatever is happening on the inside, be more of a focus. So that's what we did. This is the before, and this is the after. So, you can see we really reduced, I guess, the intensity of the cell outline. We also bumped up the size of the T helper cell here because scale-wise, I think it wasn't really accurate. We can't be accurate all the time. I mean, these proteins aren't that large in real life, but to the best of our abilities, we like to keep it pretty one-to-one, and then really the focus ends up being what's happening on the inside of the cell and also maybe the inputs and outputs of this process. So, going from before and after, we reduce the number of colors. We really thought about the specificity of the audience. They don't need to know that intensely that this is the outline of the cell. We kind of reduced that, subdued that back, and then saved the color splashes for the arrows and sort of these little vesicles.
So you can see the difference there. And again, all the same content, it just takes a little bit of tweaking. Maybe spend an hour, you know, reading and using some of these tips, and I suspect your figure will look much, much cleaner.
Last few figures here, here's another one that I thought was really nice, very clear division of three sorts of categories. So we just clean it up a little bit, thinking about areas that we can improve. So I see sort of two processes here happening, we were able to condense those into one. So here's the before, here's the after, where it was one cell leading to sort of two outcomes. I thought that was a cleaner approach. So again, there's a before, there's that after. We also incorporated these dotted lines just to give a little bit more of a segregation between these three mechanisms or processes instead of somebody accidentally assuming that they actually kind of all flow together, which could happen. It actually does look like one is sort of related to the other. In fact, they were three different. So that's what we did, we sort of created a line in between the three and then called up certain things that were a little bit lost in contrast. So again, using our grayscale checker, this little X over the protein was not that visible. So if we called it out specifically, you can see here it really looks striking that this does not bind here or there's no interaction that happens as a result. That's a lot more clear to me in this example than what's happening here. So grayscale again could be your best friend when checking for legibility.
Okay, and then let's see here. We already talked about this example so you know, reducing the number of colors, the directionality of legibility, and then finally, this is sort of a nice way to round out, you know, why I believe BioRender is so strong in helping create figures for your Grant applications is sort of the uniformity of your figure.
I know I see a lot of figures or you know scientists will come to us and say can you help me make my figure look more uniform and I'm sure you've been in this situation where you have photographs mixed with uh, you know, real data and graphs from say Graphpad Prism you've pulled them in, maybe you have you know icons in BioRender as one of your steps, and then you sort of have this like mishmash happening of different visualization, I guess modes. So photographs, hand drawings, old graphical objects that you've chopped up and pulled in.
So to avoid that, you know, I always like to try to keep things consistent by having everything look like one artist drew the whole thing and for the most part, that's usually possible in BioRender. So here's the before and here's the after where we actually just used all the same vector type images. We were able to find this machine in BioRender, I think it's called 7100, capillary. There we go. So if you look through our icon library, you'll see that we've got a huge variety of really beautiful machinery. These are all drawn in-house by our professional illustrators and they're all vector-based. So no matter what objects you pull into the canvas, it's going to look like, you know, the same artist drew it even though they're, you know, seemingly disparate concepts. So they all have that sort of really beautiful vectorized style.
So as best as possible, like I said, try to keep the style consistent, especially if you have these sort of stepwise processes or experimental flows. You don't want it to look like it was, you know, hodgepodge or mixed in together sort of a cut and paste of a bunch of sources.
Okay, and I love this little tweet that we were called out on that, you know, review article or Grant application, independent of that, you know, generally speaking, writers will think, "I'll do the figures at the end," and reviewers will say, "I'll review the figures at the start." And naturally, you know, that's going to happen because if you come across something like this, our eyes are going to naturally want to focus on the easiest thing to read, which is usually the picture.
Again, I highly recommend starting early. There are questions about when to start your figure. I started as early as possible and iterated and got feedback as you would your writing part of the grant.