Posted in Ph D posts

Writing, writing, writing…

I really need to learn how to draw on Illustrator #PhDLife #Science #Writing

As mentioned in the previous post, I’ve been a bit MIA of late due to some urgent writing business, which gave me an idea to write a post about manuscript writing.

***

So, what do I mean by “manuscript writing”? As a Ph. D student, I have a Ph. D thesis to write (an absolute requirement to obtain a Ph. D in the first place), but I can also write manuscripts to submit to a scientific journal for peer-reviewed publication. While not an absolute necessity, having some scientific papers under your belt stands you in good stead when applying for jobs- especially if you want to stay in Academia.

Simply put, manuscripts are what scientists are constantly working on for their entire career. If you can publish a manuscript in a respectable scientific journal, it means your research is of interest (it has a high impact to the field). You can use that prestige to apply for more research funding, to then do more research, which you ultimately publish… and so the cycle continues. The classic “publish or perish” phrase comes from this cycle. If you can’t publish your research, then you won’t get any funding. Publishing = money.

Obviously there are many things that are wrong within this system (that’s a whole other can of worms), but I’ll just take this moment to write an informative piece on the processes involved in getting your research out there.

***

Manuscripts, unlike a Ph. D thesis, are much shorter, and relatively brief in their introduction. The main structure is something like this:

Title: Something informative (and possibly eye catching) to draw the reader in. Something I’m not particularly good at doing (“look at meee, I’m interesting!”).

List of authors and their affiliations: This is probably where things can become political very quickly- basically, the order of the names is actually quite significant. The first person (“first author”) is the one who did the most amount of work, and the last person did the least amount. Because of the significance and prestige of being the first author, some people may end up fighting over who deserves first place. Sometimes you can have co-first authors (where numerous people are first authors), but either way, a first author paper is something to be proud of… provided you didn’t obtain it in a fraudulent manner!

Abstract: ~250 words to describe what it is that you did and why the paper is worth reading (there’s a larger explanation in my “scientific conferences” piece). I tend to leave this bit til last when writing (because it’s a summary of the entire paper- I’d rather have it all written before I go on to summarise it).

Introduction: Gives enough information to help the reader interpret your manuscript (specifically, your results and discussion). Starts fairly broad, showcasing current literature (what’s currently known and published), then highlights the missing piece of information which will segway nicely into your hypothesis (a logical statement of what you expect to happen, based on previous literature and data), aim/s and (optionally) your main findings. Writing this section can be time consuming. Even two sentences can take an hour because you’ve “gone down the rabbit hole” (ie. have 50 tabs open on your browser) to find the perfect reference to cite and support a statement.

Materials and Methods: Contains numerous subsections outlining how you did specific experiments. Generally speaking, they have enough information in there so that anyone reading it can repeat the experiment independently. Details about the specific cells or strains used, reagents and chemicals (including manufacturer’s details), all need to be included. This bit can be written with Netflix or music blaring. Very little concentration required.

Results: Also divided into subsections, this section begins with a brief reiteration of the aim/s and how you went about achieving them (brief summary of the methodology for each subsection). Results (and accompanying figures) may not necessarily be presented chronologically in the order that the experiments were performed. Instead, they tend to be presented in the order that flows best. “We wanted to answer this question, so we did these experiments, and the results are shown here, this suggests A. (Next section) previous section showed this, which suggests A, so to confirm this, we did experiment 2” or “…suggests A. (Next section) previous research into A ties into idea B, so we explored their relationship”. Main point here is to describe the results but not to delve into what it might mean. That’s for the next section… Also a section which can be written while watching Netflix, because you’re literally just making figures, or describing what you’re seeing in said figures.

Discussion: Generally starts with a brief summary of your aims and what you found (ie. remind the reader what the hell you were trying to achieve). This is where the most amount of effort should be given in any manuscript. Your results have been showcased in the previous section. Now is the time to interpret what it all means- the key point being, you need to discuss in depth why your results are significant. How does it tie into what’s currently known? Does it confirm previous findings, or bring something new to the table? Why do your results matter? A good discussion should highlight the pros and cons of your results. Highlight any limitations and explain how you could address it (“future directions”), and then finish it off with a brief summary of your main findings alongside a broader, general comment on how this research could help your field. This section takes the longest to write, in that you may find yourself thinking about thinking. Long silent moments of deep thought required. It’s good to find somewhere quiet to write this section.

Aaaand of course, then there’s an appropriate number of references in a reference list- because the paper should cite previous peer-reviewed studies when it makes a statement about anything. Scientific papers are built on previously published scientific data. If you don’t use a citation manager (I use Endnote), you’re in for a bumpy ride.

***

Depending on the journal, the methods section might be at the end (after the discussion). The journal might have other formatting instructions (font, figure formats, table formats, how to write chemical names, etc), but they’re generally listed in an “instructions for authors” document on their website. Either way, the above components are always there. There’s generally an acknowledgements to note any contributions from other people (“Thank you to XXX for antibodies/plasmid construct/cell lines” are quite common in my field), a funding statement (“XXX is supported by a XX grant”), author contributions (“XX performed majority of experiments under guidance of X”)…

Either way, once a draft is written, it gets sent to all co-authors so that they can give feedback on the manuscript. Their names are going to appear on that paper, so they deserve a say in what their name is associated with. Once all comments are ignored incorporated, you can finally submit the paper to a scientific journal of your choosing. Oh, and you do need to pay a couple thousand dollars to publish. Some places even charge for colour publishing (even though they may solely be an online journal now). Sneaky.

***

There are many different journals available, all for various different fields. In addition, scientific journals do have a tier system, and even though it’s technically not the standard any more, most people still use “impact factor” as a measure of how good/prestigious a journal is. For instance, Nature (a very prestigious journal) had an impact factor of 41.577 in 2017. At the other end of the scale, Journal of Bacteriology (a very old and still respectable journal) has an impact factor of 3.219 in 2017. People can simply see the name of the journal and quickly decide how good your research is, based on the impact factor. The higher the number, the “better” your data needs to be (sometimes it can be a bit wishy washy, though- again, can of worms), so if your data isn’t super significant to your field, maybe submit to a lower tier journal. If you think the work is solid and it’s groundbreaking stuff, aim high.

The initial screening process is done by one of the editors of the journal (sometimes a board of them). They’ll give the manuscript a quick skim, and decide then and there whether the manuscript might be worthy of publication with them. Sometimes it can get rejected right here. Maybe it’s not very good, maybe it’s not the right fit. Hopefully they tell you why.

If they do like it, though, they’ll assign and send the manuscript off to reviewers. These reviewers can be nominated by yourself (although not a guarantee that they will get picked), and are generally expert researchers of your field. Usually 2-3 people get assigned for each manuscript, and they could have anything from a few weeks to a couple months to independently and thoroughly go through the paper. They’ll be looking for gaps in the methodology, incorrect interpretations of results, inadequate discussion, incorrect terms (or so you’d hope)… They do remain anonymous, though, so when you inevitably get their comments back, they’ll simply be Reviewer #1, #2, etc. Ultimately they will decide whether your paper is worthy of publication.

Feedback can be anything from:

– This paper is awful, reject immediately (no publication). They may give you reasons why, or they may not.
– This paper is only okay, because you haven’t addressed the question in its entirety. Do more experiments here, here and here, or re-do them with these conditions added, to address it fully (chance for publication but can also be rejected)
– This paper is pretty good, but you could do this experiment/re-do this experiment with these parameters added, and it’ll be better (chance for publication)
– This paper is good, there are some minor comments mostly relating to grammar/semantics in the bulk text, or I just don’t like that colour scheme you used for Fig 2A (ideal scenario) (high chance for publication)
– This paper is amazing and no alterations are needed (extremely rare) (very high chance of publication)

Reviewer comments can be anything from constructive to downright rude. Just depends who you get. Some people might have something against one of the authors, and purposefully write awful things… it happens. Hopefully you get constructive ones, and the “concerns” are only minor. There is a weird… thing, where Reviewer #2 is always weird/bad. Not sure why. It is a thing, though.

Once the comments are back, however, and the paper doesn’t get immediately rejected, you can write a rebuttal to address the comments. You can say you’ll do or re-do experiments to address certain comments, or explain why you won’t be addressing others (maybe what they asked is physically impossible… that also happens)… either way you’ll be given a time frame to address any revisions, and you can submit a revised draft of your manuscript. Hopefully the revisions are good enough and the paper finally accepts your manuscript.

Acceptance is when you can finally pop the champagne. It may be a few weeks to a few months before it gets officially published, but if it’s accepted, you’re guaranteed a publication, so you’re good. My lab has a tradition that if a paper gets accepted, we have “paper lunch”- essentially the boss pays for us all to dine out for lunch to celebrate. It’s pretty nice.

That’s about the general gist of the publishing process in science. Hopefully it’s a relatively simply yet informative post. I’ll now get back to my Ph. D thesis writing…

Author:

A wet-lab based Bacteriology Ph. D student residing in Australia.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s