Two Triage Scores Later: Anatomy of a Grant That Finally Funded

The first summary statement arrived on a Tuesday afternoon. I read the triage verdict standing in the lab kitchen, coffee going cold, while a graduate student asked me a question about western blots that I answered on autopilot. Unscored. Our R01 application — eighteen months of planning, preliminary data, and careful writing — had not even made it to full review.

This is the story of what happened next, told in the spirit of the candid lab-keeping we try to practice in our science. Grant rejections are usually discussed in whispers, if at all. They deserve better treatment, because in the current US funding environment they are less an anomaly than a stage of the process.

Rejection One: The Numbers Problem

The first cycle's reviewer comments were, in hindsight, precise and fair. Three criticisms recurred:

The hypothesis was too broad. One reviewer wrote, with some exasperation, that we had proposed to 'elucidate the mechanism' of an entire pathway. Fair. We had bundled four aims that each could have anchored their own application.

The model system raised doubts. Reviewers questioned whether our in vitro system would translate, and we had no in vivo bridge to reassure them.

The team lacked a key expertise. We had no computational biologist on the application at a moment when the aims demanded one.

Our first instinct was the wrong one: we drafted a rebuttal in our heads, mentally defending every choice. The better move — suggested by a senior colleague who had survived four cycles of submissions — was to treat the comments as peer review of a manuscript. The reviewers were not adversaries. They were the three smartest readers our proposal would ever get, and they had told us, for free, exactly what was wrong with it.

Rejection Two: So Close, and Yet

We restructured aggressively. We cut an aim, added a co-investigator with strong computational credentials, and included preliminary in vivo data. The application scored — but just outside the payline. Percentile 21 in a year when our institute funded at the 12th percentile.

The second summary statement stung differently. We had done everything the reviewers asked, and the science was now genuinely better. The critiques this time were narrower: one reviewer remained unconvinced by a single aim's feasibility, and another wanted an alternative approach spelled out in more detail.

What Actually Changed the Outcome

For the third submission, we made a short list of revisions and, just as important, a list of things we refused to change:

• We made the riskiest aim explicitly contingent. A new sub-aim described exactly what we would do if the primary approach failed, with the fallback experiment named and budgeted. Reviewers reward a plan B.
• We rewrote the significance section for a non-specialist. We asked a colleague outside our subfield to read it and flag every sentence she didn't immediately understand. Nearly a third of the section got cut.
• We stopped overselling. Phrases like 'transformative' and 'paradigm-shifting' came out. Reviewers, it turns out, are allergic to adjectives that the data cannot support.
• We kept the structure that worked. Not every criticism requires demolition. Knowing which parts to leave alone saved us months.

The application funded at percentile 9. The notice of award arrived during a lab meeting, and the celebration lasted approximately one evening before the no-cost-extension anxieties of real grant management set in. That is another essay.

Resilience Is a Skill, Not a Temperament

Here is what I wish someone had told me at the start of my faculty career: in the United States, where NIH success rates for R01-equivalent awards hover in the low twenties or below, the expected number of submissions per funded grant is greater than one. A rejection is not evidence that you are a bad scientist. It is a data point in a noisy but informative process.

Treating resilience as a trainable skill changes how a lab operates. It means we discuss summary statements in lab meeting, including the harsh ones. It means senior trainees read our rejected drafts alongside funded ones, so they can see the revision process rather than only its polished output. It means we budget time for resubmission the way we budget time for experiments.

None of this softens the sting. The sting is real, and so is the statistical reality. The difference between labs that endure and labs that fold is rarely talent. It is whether the people in them know, plainly and without shame, that the third submission is part of the plan.