Thread by @AinslieLab

I stepped down from NIH study section NANO this week. Between NANO & the 32 other study sections I have served ad hoc, I learned many things about how to construct a good grant.

Thus, I give you a song themed thread of potentially helpful grant tips!

Simply the Best: Give quantitative proof your idea is > similar current clinical & preclinical approaches. Discuss similar 1s & why yours is >. Use clinical controls in prelim data, a table quantitatively summarizing the results of yours and others and/or highlight drawbacks.

Justify My Love: w/ valid & relevant controls in prelim data, justify all your bells & whistles. Justify why your platform is needed. Its not enough to have literature references, you also need preliminary data to support it - with indicated statistical significance.

I Walk the Line: If your grant is hovering above the payline - you need more prelim data to push it to funding. Distill down your grant to what you need to support your key variables. Make sure they are supported and that your steps to optimization are clearly outlined.

Monster Mash: We all resubmit again & again. Not all critiques are = & need to be carried over again and again w/ resubmission. Reviewers don’t see old critiques with a new submission. Edit out the responses to crap critiques when you recycle & just keep the good 1s.

Paperback Writer: You don’t need a thesaurus to write an R01. Keep language simple, succinct, & straight-forward. Yes you need to use sci jargon, but make it readable and not Hawthorne or word salad. Get someone who will critically review your grant even if in another field.

ABC: We all use acronyms - limit them & define w/ 1st use. Use common 1s. Too many acronyms = poor readability. If u don’t use the acronym much, don’t define it. Ctrl-F to determine frequency. If you define it early & use it late, be cautious. If room put table in aims page.

Why Do You Build Me Up Buttercup: Have a variable & support it. Don’t just characterize a platform…optimize it! Have a variable(s) that will do that & HAVE PRELIMINARY DATA WHICH SHOWS THAT VARIABLE CHANGES OUTCOMES. This is often the issue with the grants of new investigators.

What's So Funny 'Bout Peace, Love & Understanding: Reviewers don't know your field, but a related 1. Some intro is needed. If your grant uses AuNPs 4 vaccines, you may have a vaccine expert w/o AuNP knowledge & vice versa. Cover all your bases of understanding & characterization.

Turn the Page: Space is limited & we ref our papers to not put figs in. Reviewers have a grant pile to review & can't dig up the refs. Include most relevant pub’d data in grant over paper ref. Distill down compound figs to most relevant parts, don’t just copy paper figs.

Complicated: A simpler approach is needed. Straight-forward with 2ish variables evaluated in the grant is easier to review than the whole kitchen sink. Fishing expeditions don’t score well. Use prelim data to support your variables & rationally & clearly evaluate (optimize) them.

I Can't Get No Satisfaction: NIH review is SO not perfect & bad critiques suck. It’s satisfying to tell them off in the intro, but don't. Address nicely the critique high points (even if crazy). I use blue text to indicate major text changes, including fig captions of new figs.

All the Small Things: List groups in animal exps. Justify animal numbers & sex in res strat. Discuss stats & include stats in prelim data. If 1 aim depends on another have prelmin data to shows key aspects of the dependent aim can be done. You lose your new PI bump if you co-PI.