Description: Danielle Matsushima (Director of Research and Strategic Initiatives, Columbia University) and Shiz Aoki (Co-founder & CEO, BioRender) share tips for how to visually communicate your science in research proposals to make reading easy on the reviewers’ eyes.
This webinar was recorded at VISUALIZE 2021, a virtual BioRender event dedicated to advancing communication in science.
I'm excited to be here. I am Danielle Matsushima, the Director of Research and Strategic Initiatives at Columbia University's Vagelos College of Physicians and Surgeons. Today, Shiz and I will be presenting on how to design your grant for the reviewer and give you some tips that would make your proposal pleasing to the reviewer’s eye.
Unlike Shiz, I am not a designer but I lead a team of research development professionals who help faculty with the strategy behind creating a compelling grant proposal. Since 2016, we've worked on over 200 proposals and we've facilitated the submission of several large complex proposals, Career Development Awards, and early-stage investigative research grants, and we've brought in close to $190 million in awarded funds to the medical school. We've worked with several experienced PIs as well as first-time applicants, and our team has learned some tips along the way for grant visuals, and we'd like to share them with you today.
So, what is the goal of grants?
The goal of grants is to get the money in your hands as a scientist so that you can go and conduct your work. When you're writing research proposals, you need to convey why the scientific topic or question that you're asking is important. You also need to convince viewers that you are the expert and the right person to carry out the proposed work, but you also need to talk about how your institution has the resources to do the work you propose. Your studies need to be demonstrated to be feasible, logical, and well thought out, and you need to convey that the results of your study will make a substantial contribution to the field. And if that weren't hard enough, you have to do all of this in a very limited amount of space. So, you really need to be choosy about what information to include. As you're telling your story, you have to keep in mind who your audience is. Your audience tends to be fellow colleagues, so fellow scientists or clinician scientists, but you have to keep in mind that they're also running their own labs, and they might be giving lectures, they might have their own company on the side, they also are writing their own grants, picking up their kids from daycare, trying to stay active and healthy, as well as have their own hobbies. So, viewers are people like you, very, very busy people, and therefore you need to develop and design a proposal that captures their interest and is easy to understand and doesn't waste their time. This applies to what you write as well as what visual concepts you present. Also, keep in mind that reviewers may not be in your field, so you need to clearly convey what you're proposing, why it's important, and how society will benefit by funding the project. Clear writing and clear visuals will help you get your message across to a broad audience.
Effective visuals are extremely important. They help with a good first impression. You know that first look when the reviewer sees the grant. If it's a professional-looking grant, they may say, "Hey, this person is really detail-oriented." And if they're detail-oriented, then they might also be really detail-oriented in the work that they perform, and so you might be positively influencing the reviewer at that point. They haven't seen your science yet, but just the design can give them a good first impression. Effective visuals also allow the reviewer to see exactly what you're proposing, and this will help increase understanding. So, just the words and text may not be enough, and having a visual, having that data or having a schematic, really helps the reviewer see the same thing that you want to get across. Visuals also demonstrate feasibility so that by including your data, reviewers can determine whether or not what you're proposing is something that can be done. Including effective visuals also helps with a favorable bias towards you so that ultimately reviewers will be convinced, and they will fund your work.
So, there are four main ways to display data in a grant. The first is the use of a table. Tables are used to display lots of data that would otherwise be difficult to explain in text form. You can see here that I've taken away some of the grid lines and I've added pops of color. This keeps the table interesting, and it also helps convey order within the table. You can use color to highlight information that you really want reviewers to pay attention to and help guide their interest to where you're talking about things in your text. Another way to display data is to use a graph. Graphs are used to visually convey meaningful relationships between data, so this could be to demonstrate a pattern.
Trend or an interaction between values: There are several types of graphs. I'm not going to go through them all, but really keep in mind that you should choose the graph that depicts the relationship you want to show. So, if you want to show a comparison, use a bar or histogram. If you want to show parts of a whole, use a pie chart. One thing I do want to tell you is that some reviewers will judge the science based on the information that's present in the graphs, and in fact, sometimes this is all that they'll remember. So, make sure that you're spending time crafting your graphs in a way that is visually pleasing to the reviewer and helps them remember the salient points that you want to get across.
Another way you can display data is to use an image or a photograph. This really concretely helps the reviewers see what it is because it's the exact same thing that you want to show them. They should be self-explanatory, so always label the important items within that figure or that image. Explain any symbols you use, any arrows. Really define that in the figure legend, and if you're doing an image of this one, include scale bars.
Lastly, you can use diagrams or schematics to convey information. These are great because they're visually interesting, and you can communicate information quickly. So, including a diagram that could be explanatory to get across your entire hypothesis or the goal of your work. You could depict an experimental design, show how something works, or interactions and how events are ordered in time and space, like a timeline. The key to a good diagram is one that stays true to its purpose and really helps the audience understand what you're presenting in a very straightforward manner. Therefore, you want to keep them as simple as possible. So, with this diagram in your grant, if you're only going to talk about the left side of the pathway, don't include what's in this red-orange color. It's detracting away from the main point, so just don't even include it. Simplify as much as possible.
What makes for great grant figures? Honestly, many of the design elements that make for great presentations also make for great graphic grant figures, and we've heard a lot of these throughout the talks yesterday and today. So, these are like color, space, typography, and flow. Things to pay attention to.
First, I want to start with color. Color can be a hindrance or a tool. So, as many of you know, this is a heat map which is a visualization technique that allows you to see the magnitude of a phenomenon, often gene expression, as color in two dimensions. So, I don't think this is done anymore, but this is a red-green color scheme that's here, and for 90% of the population, this color scheme is good. However, for someone who's red-green colorblind, this is what the heat map looks like. You lose that information; you lose the ability to be able to determine what's up or down-regulated. However, if you choose a red-green colorblind-friendly scheme, this muted red-yellow and blue that's here, that information up at the high expression or low expression is maintained for people who are red-green colorblind. Okay, so this is just to demonstrate that your color choice is really impactful in terms of what the reader understands.
Color can also be used for emphasis. So, here you see that there are two time points in two different shades of blue. Since they're the same color, it suggests that there's some sort of relationship between the two time points. One can assume that the other conditions, maybe its genotype or temperature, are all the same, and what is different is just that the time point is taken. However, if I'm going to show you this graph with two almost opposing colors on the color wheel, red and blue, you automatically assume that there's a difference between the two, and indeed, red is control and blue is the mutant in this case, or conditional knockout. Here you can tell that there's a clear difference. I also want to state that humans are hardwired to like warm colors, so the red-orange and yellows more so than the cool colors, blue, greens, and purples. So, your eye is automatically navigating more to the red than the blue, and so there's competing interest about what to look at. However, if I change the red to black here, what ends up happening is that you can easily, clearly see the relationship because you kind of assume that black is baseline, and you can see what's happening in the conditional knockout. So, in cell types A and B, as well as C, you see a significant reduction in the cell number in that type. So, I just say this to keep in mind that reviewers are reading quickly. They tend to be skimmers, and so anything you can do to help simplify the take-home message really helps when trying to get your point across. Also, some reviewers still print out their grants, and they might print them out in black and white. So, when you're choosing colors, try to choose something that has high contrast and will print in grayscale.
The second design element I want to talk about is space. Space has a purpose. Last summer, these red signs popped up in all of New York City parks because they wanted to remind people that keeping six feet of distance apart reduces the spread of SARS-CoV-2. Well, blank space also has a role in grants as well. Okay, so empty space breaks up the text and allows for readability. So, you can see here on the left that this page has high text density, lots of use of bold, underline, and very few spaces where your eyes can rest. Okay, so this isn't real text. It's just the Latin template text, but let's pretend for argument's sake that we have edited this. We've reduced the redundancy. We've included clarity. What's left is you get additional space that we can put spaces between the paragraphs to allow reviewers to have a break as they're reading and assimilate all that information that you've given. We've also bullet-pointed innovation points here, and we've reduced the bold to only include this one sentence here. Perhaps this is the overall goal of the proposal. All of this adds to the ability to read this document and creates a cleaner visual here. One thing I want to mention is that justified text or this ability on the square-like looking text that's here versus left-aligned text, which is seen on the right here, tends to take reviewers longer to read justified text than left-aligned. But I've polled through the years, and reviewers tend to not care if you use one or the other, just something to keep in mind when you're choosing to justify or not.
Space is also good in figures too. Okay, so this figure here, all the subpanels are kind of lumped together. It's hard to distinguish like what's A from B or D from E. So, there's this idea of the principle of proximity so that one will assume that objects that are near each other relate to one another, and here, like everything is near each other. So, then you're like, "Oh, everything is relating to each other." What you can do or what I like to do is I like to kind of draw borders or margins between the sub-panels. So, here you can see that in gray, and then once you've removed them, then you have some nice space between the sub-panels. What this allows is for the reviewer to concentrate on one little bit of data at a time, and then they can really focus on what you're trying to get across in each one of those subpoints again I know research strategies are limited in space and this is probably asking a lot but even creating a little bit of space between the sub panels will really help reviewers take in that data and digest the overall message.
The third thing I'm going to talk about is typography. Typography can send a message and the text you choose matters. So here this text is in uppercase and if I were to make it lowercase and only bold the word "matters" the emphasis then is on the word "matters". Here, the choice of formatting sends a different message.
The type of font you use is important. It affects legibility, it affects the tone that is perceived, and it also affects the level of professionalism in your document. So keep that in mind when choosing a text. All fonts are divided into two kinds: serif font and sans serif font. Serif fonts are distinguished by these little protrusions off the characters and if you can see here in the red circles, sans serif fonts are lacking in those.
So, they have different pluses and negatives. Serif fonts are really good for small characters. They're easier to follow one line at a time and they're good for written documents, whereas sans serif fonts are seen as simple and pure. They're better to be seen from far away and they're good for slides or poster presentations. I've just told you that serif fonts are better for written documents, but I've polled some reviewers and they actually like sans serif fonts better. So things like Arial and Helvetica, they like to see the text in that. And definitely in your graphs, use sans serif font.
Also, decide on text that increases legibility. So here I've put different types of text that's been formatted differently. Sentences that are easier to read are at the top where sentences that are harder to read are at the bottom. You can choose different formatting to really create a design within your grant so that reviewers can navigate things easier. Bold is good for emphasis, the second line is good for the body of text, this third line where everything's in all caps and bold is good for the headers, and then at the bottom with this white-yellow, just don't use it, it's too hard to read. I've actually worked with PIs on grants and I call them "rainbow grants" where they want to make the word "project" red everywhere and the word "core" blue everywhere and they're referring to PIs and the PIs are green. What you end up getting is rainbow grants and reviewers can't concentrate on any one aspect. They concentrate on the word and the color rather than the science that you're trying to convey. So really keep in mind the formatting that you choose for your text.
It's also important to position information in logical order. So here in the US, we tend to read from left to right or top to bottom and so when our team makes organizational charts or interaction charts of all the different components of these large center grants, we try to keep this in mind. So this is a figure that two members of my team made and for the most part, the information is really flowing from the top to the bottom, within a core. So they just want to reiterate that. Put the information in the way that the reviewer would expect to read it. Other things to note with this figure is that it's well-balanced so both left and right and top to bottom, there's content there and it seems unified because we're using the same font as well as kind of this pastel color palette that BioRender has (a great color palette to be able to create that unity-looking thing). So I just want to point out that I know the design team would probably get a hold of this and have some extra suggestions to make it even more design-friendly, but something like this and putting it in your grants can really help, especially with these large complex center interaction grants.
And so when you get a reviewer comment that says, "You know, figure one is especially helpful in highlighting the vital interactions of the diverse elements of the overall research effort as well as the NIH and the larger research community," then we've done our job. We've used our visual to convey to the reviewers of why this center grant is needed and all the interactions that will take place.
I also want to talk about how it's so important to simplify as much as possible. So this is sort of a before on the left and then after that's here. So what I've done is I've made the control black, I have simplified the grouping line so instead of having brackets above and brackets below, I just have the below brackets listed. I've moved the legend closer to where the information is needed. I've also reduced the space footprint by doing that. I've only included the above error bars to get rid of some clutter. I've changed the font to sans serif font. I've increased the font size and I've rotated the X and Y axis labeling to be horizontal, again, easier for the reviewer to read that way. So with a few small alterations, you can help the reviewer's eyes from straining and from focusing on all different types of information and it allows them to quickly see that in this case, the cell number and cell populations A and B are reduced in the conditional knockout.
I want to tell you some common mistakes that we tend to see in our grants. So when it comes to data, reviewers often comment that there's too much data or too little data. So if you're putting too much data, it's hard for reviewers to follow along with your story. So really be picky about telling your story and only giving the relevant data. Also, we get comments that there's too little data and so often reviewers say that your study is not feasible because you're not showing enough preliminary data. So keep that in mind. Also, make sure you're presenting the data where you discuss it in the text. So if you're discussing figure one on page one, then figure one should also be on page one and not on page five. You don't want to make reviewers have extra work and to go and have to find that figure. When it comes to figure captions, the most common comment we see is that they tend to be too small. So make sure that when you print it out, you're able to clearly see that figure.
That interaction figure I showed you, that center figure that I would even put like a whole page dedicated to that figure. Okay, so make sure that you're giving the figures the amount of space that they're due. When it comes to captions, use font size nine or higher if you can. Also, make sure that your figures are high enough resolution that if a reviewer decides to, they can blow that picture up. This just comes down to sloppiness. Sometimes we see grants going in, you're running out of time, and you send your grant in. And that winds up happening is that your figures over here and your figure legend is down below, and they're not together. Okay, so the figure legend has migrated. It's an easy fix, right? Just checking that and moving it before you send the grant in. We've also seen things where figure one goes to figure six. Make sure that you're numbering and that you have the correct numbering, and just check that before you submit your grant. When it comes to formatting, I reiterated this tons, make sure you're not using too much formatting. If everything's bold, then nothing becomes important. Also, make sure you're using consistent labeling, whether it's your headers for your grant or when referring to a certain condition, such as a control or condition knockout. A condition knockout should always be that one color throughout all of your figure panels. Other comments that reviewers often say is that they're confused about the experimental design. So if you include a diagram showing your experimental method or outcome, you can get rid of that comment. Make sure you're always including statistics and make sure you're always discussing it in the figure legends. Include a timeline. Reviewers like to see how you're planning out your experiments and that everything you talk about will be feasible in the work period. Also, comply with funder guidelines and always ask for feedback. You know what you want to say, but you don't know whether or not you're achieving that goal until you ask someone to review it. Okay, so I want to leave you with this idea that your proposal is your brand. This is how you're representing yourself and your work to reviewers. So if you give them something on the left, there's a lot. It's hard to focus on anything. There's a lot going on between the bold and the underline, etc. But if you were to give them something on the right, you know it's clean. The use of colors draws your attention to certain areas, and so the visual layout and design is important because it enhances the communication of your ideas. It increases the perception of structure and logic, and it helps imply clear and organized thinking. That's what you want to create for reviewers – that your thought process is clear in terms of what you're proposing. So with that, I'm going to talk a little bit about how BioRender has helped our research community. At the beginning of 2019, we started sponsoring the institutional license, and since that time, we've seen a huge increase in the number of users that we're sponsoring. And last year, BioRender did a little survey for us, and one of the things I want to point out is that BioRender saves time. So out of those that responded, most people were taking one to four hours to make a figure, and after BioRender, most people are taking under two hours to make a figure. It's saving them time so that they could go be writing their grants or conducting experiments, etc. For those polled, they said that strong visuals really helped them get publications into grants. So that just speaks to the fact that including visuals and doing them well is really important to getting your grant. With that, I'm going to pass it over to Shiz.
BioRender Specific Tips
Thank you so much, Danielle, for those tips. I think I learned a whole bunch of new tips, so thank you so much. Super tactical. And let me know if my audio is okay. Give me a thumbs up or down backstage. Okay, great. So hopefully, you can see my screen. I think Danielle laid a really amazing foundation for all the considerations, including fonts, design, content, data, and if you're using BioRender specifically to make your figures for grants. I'll summarize with a couple of tips here. I've got, let's see, I'm on the clock for about five minutes, so let's run through some, I guess, BioRender-specific tips that you can implement based on what Danielle has just shared. And a little meta, I'm going to be using BioRender to present, so that's why I've got my slides panel open there. Let's see here. Maybe I'll go into full-screen mode, actually, so you can see a little bit more of the application. There we go. Okay, so when we look at the composition of any figure, left to right, up to down is generally the rule of thumb. That's how most of us read the alphabet or figures. It's the way gravity falls. So try to stick with one or two of these. Sometimes what I see is figures will be read left to right, down to up. It forks, but it goes the other way. For grants specifically, you're going to want to tell them, hey, this is what I'm going to do with the money you're going to give me. Maybe here's two or three specific aims. So I think the fork composition is quite common. I've just crossed out the M and the L because that's sort of the common poster format that a lot of people use. The M shape, you know, intro, methods, discussion, all that. Here it is in practice. I'm going to have a little bit of fun here. This is one of my favorite artists. His name is Daniel Keys. But being in the painting and oil painting industry for a couple of decades now, I'm noticing a really interesting pattern where our eyes are just attracted to the highest saturation, highest contrast areas of a composition. So here in this case, your eye probably went straight to this area, and then orange and blue are complementary colors. They're the furthest away from each other on the color wheel. So our eyes tend to naturally gravitate towards areas where there are high concentrations of complementary colors.
Near each other, blue and orange. If they're close together, for some reason, we just really like to look at it. You know, Kraft Dinner, the box is bright orange and bright blue, for some reason our eyes like that. Same with red and green and orange, yellow and purple, but I would avoid red and green for the colorblind folks. If you swap this image to black and white again, the highest contrast area (so take color out of the equation) is this sort of area here, so your eye is going to want to go in this sort of triangular composition. Here it is in practice. I'm going to show you the exact same pathway four times, but where I've focused the color and contrast is actually where your eye will go, and it's kind of astonishing. So here it is at the top. I mean, I don't even have to really tell you where to look because the color and the contrast is going to direct your eye to where to focus. So you see how powerful that is? Where your eye probably went from top down just based on the areas that I've chosen to add the most saturation and the most contrast. So keep that in mind. Again, going back to our little cheat sheet here, you know, reduce the number of colors you're using because you want to select areas of focus as the brightest and highest saturation areas. Okay, how am I doing on time? I'll speed through the next few. As Danielle said, space is your friend. So what I like to do is kind of get a good gut check of the areas that are the sort of breathing space for your figure. It's not wasted space, it's really important to have those as sort of visual hallways between your figures, your diagrams, your titles within your titles within boxes. Try to keep a consistent space. The purple's there just to show you where I have the white space, and it's quite a bit as you can see, but it's time-space well invested. A quick tip on color temperature: so as Danielle alluded to, warmer colors your eyes tend to go towards, but also it kind of connotes a sort of a villainous experience. So, like cancer cells should usually be red and lymphocytes and sort of the good guys in the body, the protagonists of your story, should usually be cool colors. So look at the drastic difference between these two where on the right, the cancer cell looks more evil, on the left, the lymphocytes look more evil. And that's not the story you want to tell, maybe it is if it's an inflammatory experience or process and inflammation is actually what's causing the negative effects in the body. So you have to kind of pick and choose where you're using warm versus cool colors. Okay, let's see here, one quick gut check you should always do at the end of every figure, regardless of whether it's a grant, publication, poster, is to swap everything to grayscale. Now, it feels kind of silly to do that, but even with BioRender we use grayscale checker to make sure we have enough contrast in our figure. Color is very distracting, so sometimes you don't realize that what you've ended up with is a really low-contrast figure, and by eliminating color as a variable, you'll notice right away that you know these cells in this tube are way too light. So what I can do is make that maybe a darker value, a darker shade, like a dark green, for example. Anyway, you can play around with that and make sure that you have enough contrast, but our little black and white contrast tracker is really good for that, so keep that in mind. And then last tip here, if I go back to my little cheat sheet, is to sort of play with transparencies. I know sometimes we're afraid to make things recede into the background, we want everything to be bold, but as Danielle alluded to, if everything's bold, then nothing's important. So if you look at this, and you squint, and actually why don't we use the grayscale checker, I can't really see what the main story is. It's actually supposed to be the lymph nodes, the tumor size, the metastasis of the cancer, and I'm not seeing that because everything is this sort of dark gray wash. So what I'm going to do is actually reduce the transparency, or rather increase the transparency, reduce the opacity of the lungs because your audience, I suspect they're going to know what a lung is, right? They know basic anatomy, hopefully, at least if that is your audience. So they don't really need to be reminded like, "Hey, you're looking at the lung." You're constantly reminding them of that, and look how much the focal point pops. So anyway, this is another common mistake I see in a lot of grant figures, is that you just really want to have the reviewer focus on the main areas. In this case, it was the tumor and the lymph nodes. And I'll leave you with this quote. I found it on Twitter last year, I thought it was great. Someone tweeted that writers say, I'll do the figures at the end, reviewers say, I'll review the figures at the start, and maybe the only thing they review, unfortunately in some cases. So, take time to make your figures as much as you do the writing.