Here are my top 9 tips for writing an NIH F99/K00 fellowship application. [I’ll update once I think of a 10th tip.] These tips generally also apply to other types of fellowship grants (e.g., F31 and F32), though my specific experience has been with writing an NIH-NIA F99/K00 grant.
1. Seek out examples.
The first key to writing an NIH grant of any kind is to understand the game. Know the NIH format. Understand the lingo and jargon. As I described here, NIH grants involve many, many pieces–much more than just your proposed research and training. Ask your PI or other mentors if you can read their previous grants (even if these are “faculty” grants like R series grants). Ask others in your department or university if you can read their NIH fellowship grants (successful or unsuccessful — both are helpful). Best case scenario, ask for previous grants and the reviewer feedback they received. Reading a grant followed by its reviewer feedback gives you a much deeper understanding of the game–of what the reviewers read (and didn’t read), how they interpreted the writing, whether they liked the figures, their understanding of the main message, and so on.
If you need an F99/K00 grant example, please contact me (kehupfeld@gmail.com). I am happy to send you my (successful) NIA grant, plus my reviewer feedback. Also check out Dr. Jean Fan’s website for their example (successful) F99 grant from several years back. It doesn’t matter if your field is different — reading examples from any discipline still helps you to start strategizing.
Also, for examples of other types of NIH grants, check out NIAID’s website. I’ve yet to see this from other NIH institutes, but for a number of years now the National Institute of Allergy and Infectious Disease (NIAID) has maintained a repository of sample applications and reviewer statements.
2. Start early, and seek feedback from a wide audience.
Try your best to start early. As early on as possible in your PhD, convey your interest in NIH F series grants to your advisor. They can help you plan accordingly. Depending on their funding situation, you may need to seek out additional co-mentors, and this is best done early — so that you can form a working relationship with potential co-mentors and develop your projects with them from the start.
On this similar note, seek feedback from anyone who will read your grant. Obviously, seek primary feedback from your main advisor and any other mentors. However — particularly if your mentors do not have ample experience mentoring successful NIH training fellowship applications — seek the advice of others who do. Ask peers in your lab or department to read your application*. Ask around or do internet searches to determine if others in your department, college, or university have received an F99 (or an F31 or F32). Ask those folks for advice. Best case, also ask if they’ll read your application.
Try to get feedback from as many people as possible. You want to make sure that your application is readable by a wide enough audience. While your primary reviewers at NIH should be relatively related to your discipline, they’re not going to be a perfect match. They may have a research focus related to one of your aims, but not the others. Or, they may generally have similar research but be unfamiliar with your specific methods. You want a wide audience to be able to understand your main research and training goals.
*By “application” I mean your Specific Aims, Research Strategy, and Training Plan. I wouldn’t worry about getting feedback from anyone other than your primary mentors on other pieces of the application package.
3. Select excellent (and funded) mentors.
Your mentors are critical to this process. An F99 application requires a decent amount of work from them too — in helping you edit your writing, and in working on their pieces. You need their full support before undertaking a F99 application.
On this note, you will more than likely be criticized by reviewers if you lack experienced and well-funded mentors. At a minimum, you probably want one (or more) faculty members who has graduated multiple doctoral students and sent multiple postdocs on to successful academic positions. Ideally, you want one or more faculty members who has mentored successful NIH fellowship grants (e.g., F or K series grants). If your primary mentor is a “new investigator”, i.e., an assistant professor without an extensive training history, you very likely want to also include someone more senior on your application. If this is the case, there is probably someone in your department who would be willing to serve as a co-mentor.
While I think this is a bit silly, it is the reality. I’ve heard of numerous peers having their F application rejected with the primary reason being that their PI is an assistant professor. It sometimes does not necessarily matter how successful this assistant professor has been, the grants they’ve received, etc. — one or more reviewers will likely dock your score unless you include someone more senior.
Case in point, my application included two senior PIs (both well-funded full professors) and, as a co-mentor, one assistant professor. I received the following [conflicting] feedback:
Reviewer 1 - praise for all 3 mentors: "The F99 sponsor...is an expert in the neural and behavioral mechanisms of motor control and rehabilitation strategies. The applicant has also assembled an excellent mentoring team with complementary expertise in all areas required to support her research training."
Reviewer 2 - praise for all 3 mentors, but acknowledgement of the new guy: "F99 sponsor and co-sponsors are outstanding. Drs. [Senior PI #1] and [Senior PI #2]...have more than extensive experience mentoring students. They are well-funded, and internationally well-respected in the research community. Dr. [Assistant Prof] is at a junior level but very highly respected..."
Reviewer 3 - gave me a worse score because of the Asst Prof: "Additional co-sponsors include Dr. [Assistant Prof], who is a relatively new Assistant Professor (2016) but with technical expertise to mentor the applicant in MRS. He is currently funded by a K award so this may pose some challenges in mentoring on professional development, career, etc., but not in the technical skill training."
So, in my case, even though I had two [very] senior PIs as mentors, 2 out of 3 reviewers noted that I included an assistant professor on the application. The third reviewer felt that this assistant professor (despite their successful K01 application and securing of a faculty position at a competitive research-intensive institution) could definitely not provide any useful professional development advice. Can you tell I think this is an unfair reviewer comment? 😉
Anyhow, until NIH revises its review policies and reviewers change their thinking on assistant professors, this is the reality. Do not write a NIH fellowship application without one or more senior faculty as mentors.
On the topic of mentor selection, also do not write a fellowship application without mentor funding. NIH fellowship applications do not give you any* significant research funding. You need your mentors’ money to do the research. This means that your research aims need to propose, for instance, analyzing secondary data collected by one of your mentor’s grants. Maybe your mentor has NIH funding that includes MRI scans; their grant aims primarily to analyze brain structure, but they’ve also collected brain function data. You could propose to analyze those data. Or, maybe you can extract novel outcome metrics from their data. Or, maybe they have enough start-up or other flexible funds to pay for you to do your own study [this described my super-fortunate situation.] Or, maybe you don’t need any money to do your work because you’re applying machine learning algorithms to “big” public datasets.
Whatever the case, you need to be very clear that you (well, your mentors) have the money to do your proposed research. Funding availability should be explicitly explained in your mentors’ statement and support letter.
Previously, I’ve heard the advice that one of your mentors needs current NIH funding for their trainees to get F grants funded. I’m not sure if this is still the advice — but, for instance, my PI had initially told me I would be less competitive for an F31 because she lacked active NIH funding (but was *very* well funded through other organizations, like NASA and NSF, and had start-up funding which could easily cover my proposed project). Again this seems quite silly — if the applicant and mentors can demonstrate that the proposed research can be paid for, why does it matter where the money is coming from? …particularly when you’re supposed to be proposing your “own” research, not simply that you’ll carry out one of your mentor’s grant aims.
Anyhow… having mentors with historic and current NIH funding is likely going to boost your score. I did not receive any comments specific to mentor NIH funding, but all three of my doctoral mentors did have active NIH funding at the time of submission — so this probably helped me some…
[*Technically, they do give you a tiny amount of flexible money that could be used for research costs, but my advice is to avoid writing that you need this money to conduct your research.]4. Clearly describe how your work is different from your mentor’s grant(s).
As I alluded to above, you don’t want to propose to carry out one or more aims of your mentors’ grants. You’re supposed to propose something original. Again, this is a bit silly given that you’re not allotted any money for research… so you’ve got to make do with what’s possible given your mentors’ funding landscape.
For instance, you could propose to apply novel post-processing methods to data already being collected for one of their grants. You could form one aim around working with a “big” public (aka free) dataset. You could conduct testing on “free” human participants, like undergraduates who can be compensated with extra credit instead of money.
Or, you could get lucky and have a mentor with enough flexible money (e.g., start-up funds) to cover your big ideas. Or, you could look for institutional or organizational funding sources to cover your work — e.g., I applied [unsuccessfully] for several other grants that would’ve actually funded my research (rather than my stipend): one through UF’s Pepper Center and another through an EMG company.
Regardless of the funding source, you cannot propose your mentors’ aims. I’ve seen many folks’ F grant scores get docked for proposing aims that too closely mirror those of their PI. You need to propose something different. And consider including blunt statements in your application stating something to the effect of, “These aims are distinct from those proposed in my mentor’s original R01 grant.” You want reviewers to know that you’re doing something new and different.
5. Bring a biostatistician on board.
In thinking about mentors, also consider how you can get a biostatistician involved. Your university may have a system for this — e.g., free biostats consulting for grant applications (particularly for trainee applications). Or, you may need to pay someone (which is also fair!). Either way, consider if you can get someone on board.
In my case, we had a biostats person consult on my F99 power analysis and stats plan. They then provided a letter of support I included in my application. My reviewers liked this.
Reviewer 1: Great to see support and help for statistical analysis.
So, if possible, including a biostats person will only help your application (and the accuracy of your stats!).
6. Sell your training plan.
Make sure you “sell” why you need training. Hit the fine balance between: 1) telling reviewers that you’re an amazing trainee, full of promise to make numerous vertical advancements in your scientific field, and 2) conveying to reviewers that you are a work in progress who still requires a decent amount of formal training. This is a difficult balance to strike, but it is critical.
The training plan is a very important piece of the application. Determine concrete areas where you need more training (e.g., does your dissertation involve fMRI, but you’ve yet to publish an fMRI paper?). From these areas, formulate specific training goals (e.g., become proficient in fMRI collection and analysis). And lastly, devise specific steps you can take towards reaching these goals (e.g., can you attend any fMRI workshops? Are there relevant fMRI courses at your institution? Which mentor(s) can oversee your fMRI training?).
Note as well that for the F99/K00 application, you’ll need to include both “F” (doctoral) and “K” (postdoctoral) training goals. Make sure that your F and K training goals do not overlap too much. This is an area where my application was criticized. You want to convince the reviewers that you need six years of training — two more doctoral years, plus four postdoctoral years. You’ll have a lot of time in those four postdoctoral years, so tell the reviewers that you’ve thought through how you will use them productively to learn new skills and position yourself for an independent investigator position.
Also note that you’ll likely need to do some digging to figure out what training opportunities might exist at your postdoctoral institution. Ideally, ask for an example F or K application from your proposed postdoc mentor’s lab as a starting point.
7. Highlight your prior awards.
Tell the reviewers (probably more than once) if you’ve received any prestigious awards as a graduate student. These are [for better or for worse], very important.
Summary statement: The applicant's academic performance is excellent; she is the recipient of several awards including an NSF graduate fellowship."
Reviewer 1: The applicant already has five published papers and a fellowship from NSF.
Reviewer 1: She has several awards and a graduate fellowship from NSF.
Reviewer 2: The applicant is well-trained academically...and received an NSF graduate fellowship.
Reviewer 2: Career-appropriate awards (NSF 3-year graduate award)...
My point here is not to brag (did you hear that I got an NSF fellowship?!) — but instead to convey the importance of highlighting your previous prestigious awards. In an early draft, I had only mentioned my NSF in my biosketch awards list (because, why is an application I wrote as a senior in undergrad that relevant to my potential success as a senior PhD student, postdoc, and future independent investigator?!). However, one of my committee members suggested that I highlight this in more places (e.g., in the “Applicant’s Background” section of the training plan), and I’m glad that I did. This seemed to be very important to the reviewers.
So, if you’ve received any of the “big-time” fellowships (e.g., NSF GRFP, DoD NSDEG, Ford, NIH F31, etc.), make sure that you highlight these. In addition, make sure you highlight (and explain) if you’ve received any awards that were major to you, but might be less well-known by NIH reviewers (e.g., a big award at your university, or a big award in a country outside of the U.S.).
8. Make it visually appealing.
I’d be the first to admit that I’m not the best at this! I have a tendency to bold/underline/italicize everything I possibly can. [You should see my lecture notes from undergrad.] I also have a tendency to [attempt to] cram *way* too much into small spaces. Don’t do these things. Use bold text, but use it sparingly. For instance, bold your aims. Bold your NSF award. Don’t bold/underline/italicize every time you try to explain why your research is significant.
Consider using boxes — e.g., I’ve seen boxes do great things in highlighting specific aims.
Make sure every figure is useful. Figures should clearly convey a message — e.g., use conceptual figures to convey your methods approach or your hypotheses. Make sure that data figures cleanly and clearly convey your point — use plenty of labels, arrows, etc. if needed. Use white space strategically in figures; have enough white space to separate out different sections of your figures, but don’t let white space take over. If needed, edit your figures in Photoshop, PowerPoint, etc. to get rid of wasteful white space (if e.g., you can’t figure out how to get a stats program to print your data plots close enough together).
In addition to making sure that your writing to clearly conveys your science, your proposed training, and your credentials, you want reviewers to be able to see these things from a quick look through your document.
9. If possible, plan to apply >1 time.
Ideally, apply early. For instance, try to apply as a third year graduate student if you can. This would then likely give you time to revise and resubmit your application if it’s not funded the first time. I applied as a fourth-year PhD student. If I were to do it again, I would’ve applied as a third year [though, in my case this was not possible, because that fourth year of my PhD was the first year that NIA offered the F99 program].
10. TBD.
That’s all I can think of for now! I will update this post whenever I can think of a 10th tip 🙂
Caveat to everything: Please note that I am writing this series of posts based on my experiences submitting an NIA F99/K00 application in the Fall of 2019. Should you apply, please make sure that you refer to your respective RFA, institute guidelines, etc – as these can change year to year. Also please note that these posts contain my personal [unfiltered] advice. Refer to your PI, university, program officer, etc. for more formal advice on these topics.