#ritaunsolicited thread 2/2 on abstracts for methods papers begins here! Please find 1/2 first if you're interested, or this will make almost no sense :D

One wants to set up the work, without being too broad. I avoid “Decades of work have revealed…” type statements in abstracts.

One thing I always do when editing is just skip the first line entirely and see if that actually changes my understanding of the rest. If not, it might be too generic and those words could probably be put to better purpose elsewhere.

Switching gears a bit now. When I was in grad school, on the final exam of MolBio II there was a question where we had to reconstruct what happened in a paper based on the abstract. What experiments they did, what controls, etc.

The abstract was from what we at Methods lovingly call a ‘results paper’, and we had to reconstruct the methodology, validation, controls, etc just from the abstract.

I still remember that because it was such a fun way to reverse engineer a scientific story using techniques we had been taught in that class. It was only possible because they chose an excellent abstract for us. (If my memory serves, they did ChIP-seq).

That is another ‘big picture’ way I look at abstracts and whether they are going to inform readers about the paper that follows.

I wanted to end with just some examples of our papers that I thought have excellent abstracts.

First is a paper I handled, that had the extra challenge of being only 80 words. All credit here goes to the authors; I don’t think I made any changes to it during line editing. Just a spare, clear, but complete summary of the work. https://www.nature.com/articles/s41592-021-01341-x…

I think this one has a nearly perfect first sentence. It’s long, but it tells the reader why they should care. And the rest nicely encapsulates the study. This paper was handled by Nina Vogt. https://www.nature.com/articles/s41592-021-01390-2…

I like this one as well. https://www.nature.com/articles/s41592-021-01307-z… It sets the stage, names the method, says how it works, describes why performance gains matter, and then tells you they’ve made the effort to make a user-friendly tool to implement the method. Arunima Singh handled this paper!

And that’s the end of this very long thread (2/2). Thanks for reading! I hope these tips and observations are helpful, especially for early career researchers. I am happy to answer questions! Please feel free to share your advice or your favorite abstracts!