Submitting My First Paper for Publication

As someone who came straight from undergrad into a PhD, it is safe to say that I knew very little about publishing papers when it came time to submit my first research paper. Yes, I managed to get my name on a paper published at a conference through my undergraduate research (shout out to both my wonderful grad student and the lab research scientists). But I was fairly detached from the logistical part of the publication process and very minimally exposed to the world of journals and peer review and impact factors.

While I can’t pretend to be some sort of expert, I have now gone through the process of submitting to PNAS (acronym: Proceedings of the National Academy of Sciences, pronunciation: pee-nass?). I also had some talks with my advisor about the value of publishing and what it really means for your career. My thoughts on that and my advice for submitting to PNAS are below.

What does publication really mean?

When discussing what topic to tackle after I had written my first paper (which was funded by a grant outlining an area of inquiry for me), my advisor said some things that really stood out to me. He said that some of the earlier papers he’d worked on in his career, which were more focused on niche areas of math and theory, were never really brought up again after he published them (I think I support his point by being a PhD student of his who hasn’t read them). No one really asked about them, and he didn’t think about them much after publication. He said he didn’t want to work on papers like that now; he wanted each paper to really matter and be important. Despite me sometimes wanting my first paper to disappear into the LaTex cloud (is there such a thing?), I couldn’t agree more. I want the research I do to have a lasting impact and to answer questions that people really want to know the answer to.

In a similar vein, publication in academia can definitely start to feel like a game of quantity over quality—I hadn’t heard of it till fairly recently, but the phrase “publish or parish” seems to describe the situation pretty well. (I am purposely not diving into the recent high-profile stories of academic dishonesty enacted by professors at very prestigious universities–far more skilled writers/intervieweres/investigators have covered that (please see Stephen Dubner and the Datacolada team)). As someone who has not and does not plan to go on the purely academic job market, I can’t say I’ve experienced this directly, but, even in my PhD program, you earn the degree once you’ve written three papers which can be compiled into a dissertation. While the papers do not all have to be published, I can say that based on the work that went into my first paper, I want the reward of having all my future papers be published and available to people interested in understanding my work and my capabilities as a researcher. So, at this time, I fully plan to publish three papers.

To be clear, I don’t think measuring the quality of a researcher by publication count (or the more refined h-index) is a crazy idea. But like admissions to undergraduate and graduate programs, you’d hope that in academia, where people are being hired to act as both teachers and researchers, a more holistic approach would be taken than just looking at someone’s research performance. As far as I have been able to tell, a semester of teaching from my favorite professors has impacted me more directly than any one of their publications ever will.

Submitting to PNAS

Is your paper of “broad interest”?

I think it’s fair to say that PNAS has a unique focus as compared to what I’ve heard called “trade journals”. Whereas these trade journals are very focused on a specific subject or industry (one example might be the journal Analytical Methods in Accident Research, which publishes papers that advance statistical methods for studying vehicle crashes), PNAS publishes papers of “broad interest”. I like to think of broad interest as meaning the results of the paper could be interesting to anyone, regardless of their expertise. One consequence of this is that if you write a paper whose findings are important, but not necessarily revolutionary or interesting to people outside your field, it’s less likely to get accepted to PNAS¹. For example, if you set out to test the validity of a long-used method for detecting breast cancer and your finding is that the method works well, maybe that gets published in a trade journal. If your finding was instead that the method doesn’t work well for women of certain demographics and has resulted in under-diagnosis in this group for years, you’d be more likely to get accepted to PNAS.

Styling your paper for PNAS

If you feel your paper is a good topic/findings fit for PNAS, then you should definitely submit there. I’ll preface everything I’m about to say with the fact that PNAS is format-neutral at submission, meaning they have no requirements for the format of your paper, except that it includes a title page, abstract, significance statement (this is a unique thing to PNAS), main text, references, and an SI as applicable. However, if you’re interested in playing mind games with reviewers/PNAS member editors (subtly influencing them with your very limited power), I think submitting your paper with the following things implemented can help your chances:

1. trimmed down to their preferred length (6 pages, but they’ll accept up to 12)
2. with the content formatted in a broad-interest-style
3. using the PNAS LaTex template or Word template [lowest time required]

As you may have guessed, implementing those three style suggestions takes time, and while I wish I had enough data to construct a pareto frontier quantifying the time invested in styling vs. likelihood of acceptance tradeoffs, I do not. So take what you want from this list based on what you think is worth it.

Elaborating slightly more on 2): PNAS papers tend to have relatively short methods sections—usually a page or less (though of course it depends on the field and methodology used). The focus of the papers, in terms of how they are written for PNAS, tends to be more on the results and discussion, which makes sense given the broad interest nature of the journal. This of course connects to 3); when you are looking to cut content or move things to the SI, I think it makes the most sense in this case to move methodological details.