The Art of Science Writing, Part 2: Putting words onto paper (or, more likely, Word documents)
Last week we began our how-to-guide with the key steps that need to be taken before you start writing a manuscript. We stressed the importance of reading, both the scientific literature relevant for your field, as well as the benefits of personal reading outside of science. Reading ensures that you have key information fresh in your mind, and also shows you how other people write and construct a story. You will be the one that makes your own unique manuscript, but other manuscripts can show you what a finished product looks like in terms of organization and structure. We also went into some detail on making an outline, or if you prefer, a storyboard. This provides the framework you will build off as you start putting your story together.
Before jumping into the five steps for writing manuscripts, I wanted to touch briefly on your writing environment. I’ve heard some people say that they can only write in a certain setting, that they write better at home or in the office or in a sound-proof room, or that they have specific needs in order to get writing done (e.g., loud music, complete silence, endless coffee, bottomless pretzels, and really anything in between). It’s good to have a process in place or a tool that can help you write, but be cautious of getting stuck in the mindset of feeling like you can only write under certain conditions.
There will be times in your day or your week when you’ll have some downtime, whether it’s 15 minutes or an hour between running experiments or going to meetings. If you’re thinking about ideas for a manuscript, write them down as they come. Even if it’s a paragraph that you only end up using a couple of sentences from, it’s important to get these ideas out there in a tangible form so you can rearrange and polish them latter. Writing is one of the most important parts of being a scientist. It documents both your thoughts and your hard work and transforms them into a story someone else can learn from-so preparing yourself to be ready to write at any time and in a variety of settings is an important career skill.
Step 1: What’s the story, morning glory?
Going back to Step 0, what do you have at this point? You have a detailed story board/outline of the relevant literature in your field, you have your figures in a mostly finished state…now what? Before you start taking that story board apart and fitting the ideas into text, write the last paragraph of your introduction. In our last post we mentioned that this paragraph describes the ‘Aim of paper, experimental objectives, and also list any specific hypotheses.’
But why do we start here? This is the core of your story: what you’re doing, how you did it, and what you thought you’d get. From a more philosophical viewpoint, this is also a key part of the scientific method, showing the progress between ideas and knowledge and how you use your work to generate new information to shed light on something not known before.
To see this in action, I’ve included the last paper from my PhD, which ended up being one of my personal favorite papers, partly because of lessons learned the hard way in the first two papers. I’ve highlighted the key areas:
'The objective of this study was to evaluate changes in gene expression coupled with in vitro nuclear receptor assays to evaluate the androgenicity of water downstream of the paper mill on the Fenholloway River. Two specific aims were developed: (1) evaluate mRNA levels of vtg, 17βhsd3, and zp2 in the liver, shh in the anal fin, and global hepatic gene expression profiles associated with paper mill exposure, and (2) determine if chemicals in the Fenholloway River could bind to the ligand binding domain of androgen and progesterone receptors. We hypothesized that modulations in gene expression patterns and in vitro analyses would be indicative of androgen exposure and that global gene expression analysis via microarrays would provide insights into the mode(s) of actions of the chemicals present in the effluent.'
The study wasn’t a complicated one, and I strove for clarity and simplicity in how I developed this paragraph. Work on this paragraph before any other part of the paper and have your PI or another graduate mentor review it for you. Then once you’ve got them on board with your idea, print it and keep it off to the side to remind you to focus around this core of the paper. Use this paragraph as a framework for your manuscript. As you write, you should be considering how to address the hypothesis/hypotheses using your specific aims and project objectives.
Step 2: Start from the middle
Once you have the last paragraph of the introduction, you’ll actually want to go to the middle part of the paper next. In the case of your manuscript, the introduction is the beginning of the story, the methods/results is the middle, and the discussion is the end. So before jumping back into the introduction, finish the figure captions and write the materials and methods section (as an added bonus, these are also the two easiest parts of the paper to write). A methods section is essentially structuring your lab protocols and procedures into a narrative form—keeping the most relevant parts in the narrative and citing other papers/protocols to keep the section from becoming too long. Writing this easier section first can help you get into the writing ‘mood’ and can also remind you of exactly what you did in the lab before you write about it.
For the results section, keep this to a very cut-and-dry overview of what each figure depicts. This part of the paper shouldn’t include data interpretation, just evaluation. As you’re writing these middle sections, go back to your specific aims and hypotheses and see what the data say about them. Work on these questions and use them to help guide what you say in your results section and also to frame what you’ll bring up in the discussion:
Step 3: Set the scene
Now we’re ready to move to the introduction. As we said in our previous post as well as our perfect presentations post, the format of the introduction is presenting a specific problem, its overall importance, and your approach to solving it. We also talked last week about how the outline can look for the introduction (and you already have the last paragraph, so we took that one out):
- Paragraph 1: What is the problem and why should the reader worry/care about it?
- Paragraph 2 (and maybe 3): What’s been done to address/know more about the problem so far
- Paragraph 3/4: Knowledge or tools that can be used to further address the problem
With your outline already sorted, you should be able to fill in a few sentences about each idea. The first paragraph should give a short overview of the problem at hand, including definitions and explanations of key concepts in your research area. This is especially important for people outside your field—those who work in this area will likely skip over this part of your paper, but someone unfamiliar with the tools and concepts you’re looking at will need to get a big picture understanding of your work in a single paragraph. For example, if your work is looking at Gene X and its role in the immune system and how it impacts cancer drug effectiveness, you don’t need to give a broad overview of how the immune system works, but someone coming from the field of neurobiology should be able to understand the basics of what type of study system you’re using and why it’s of relevance for your work.
The second (and potentially also the third) paragraph will be more of a short literature review, which you can expand on more in the discussion as needed. Avoid dumping all of the existing ideas or possibly relevant literature in this section, since it will make it an unreadable series of facts. Start by simply asking ‘Who else is working on a similar topic to mine?’ and work out from there. You don’t need to cover everything slightly related, but for example of Gene X immune system-cancer drug cross-talk, you can summarize the current basis of knowledge for other genes that related to system-drug cross-talk and how your gene emerged as a potential candidate for further study. The length of this section will depend on you, your PI, and also the publisher, if they happen to have limits on the total word count or a word count per section. If it’s on a total word count basis, keep this section shorter and use your words in more important sections such as the discussion.
Step 4: Bring it all together
If the last paragraph is where you start writing for your introduction, the first paragraph of the discussion is where you start writing for this last section (confused already?). This leading paragraph of your discussion is what’s going to set up this crucial section of your paper and tie your new results and previous results all together. In this first paragraph, go back to your specific aims and hypotheses. Describe what you found out through the study in the context of your initial hypotheses, and give a step-by-step overview of what you just presented in the paper. Going back to my PLOS one paper, here’s how the discussion section started out:
'We found that masculinization of female G. holbrooki continues to occur in the Fenholloway River. Paper mill effluent exposure is associated with both anal fin elongation as well as with significantly increased bone segment formation at this site. Additionally, we found an increase in the mRNA levels of vtg, zp2, 17βhsd3, and shh in Fenholloway River G. holbrooki. Through comparison of hepatic gene expression patterns to data from laboratory exposures, we found that paper mill effluent exposure resulted in an increase of genes associated with metabolic pathways, with 62 genes similarly expressed by G. holbrooki exposed to androgens, indicating a similarity between impacts at the molecular level between paper mill and androgen exposure. We also found detectable levels of both AR and PR ligands in the transactivation assay in concentrated water samples collected from both the paper mill impacted and reference sites.'
This opening paragraph can set you up for the rest of the discussion very easily, as you’ll have essentially listed out a topic for each following paragraph in the discussion. In each paragraph, think about how the results you saw fit in with key experiments from the literature and try to connect the two. What proposed pathways or models exist to explain both your results and data already in the literature? What potential ideas could explain discrepancies between your findings and a similar study by another group? It’s in this section that you’ll need to put the most work, which is why it should be saved for the almost last bit of writing.
As with writing anything, though, the one thing you don’t need to do is get it perfect the first time. The discussion is generally the hardest section to write because it requires synthesizing all the results as well as developing new ideas and explanations for what you found. Trying to put this all into writing is not an easy task-but one that you should still give a go anyways. If you feel stuck, try to go one paragraph at a time and send that paragraph to a colleague or mentor to review. Get some feedback from them as to if you’re on the right track, if your scientific logic has any holes, or if there’s a different way you can structure your arguments. The best way to learn how to write is to try, and then try some more-so if anything, don’t be afraid to put words on paper and see how it goes!
Also, don’t be afraid of a discussion that goes too long, at least in the pre-submission stages. You can always cut back, and your paper co-authors will likely also have ideas of what should go where and what’s relevant, so feel free to send them a lot and let them cut back as need be. While you as the lead author will do the bulk of the work, don’t be afraid to ask a co-author for additional editorial guidance, especially if they have good paper writing experience.
Step 5: Tie up the loose ends
While you’ll probably have to come back to your paper after your initial few drafts after your co-authors take a look, there are other things you should make sure are good to go before you finally click ‘submit’.
Literature cited: Main hint here? Use a reference tool! If you have access to EndNote then there is a very easy-to-use plug-in; if not there are other free platforms (such as Mendeley) you can use which also have Microsoft Word plug-ins. Whether it’s a long or short paper, regardless of how many references you end up having, using a reference tool will take the tedium out of this section, and will also ensure that everything’s cited in the correct format.
Tables and figures: Each journal should have a guide for authors which will specify the types of files supported and any minimum compression sizes/methods for figures. Remember that these are the part of your paper that people will often look at first-so make sure they are clear, accurate, readable, of a high technical quality, and, of course, stylish.
Acknowledgements: Be sure to thank any lab mates, technicians, or colleagues who helped out with the project but who didn’t do enough work to make it to the author list. If you have co-author who works in a company or government institution, they will likely have to include wording to reflect that this paper doesn’t reflect the companies views (they will probably add it themselves but you can make a note to ensure that they included it). And don’t forget the funding agencies who sponsored your soon-to-be published study!
The key thing to remember about writing is that you won’t get it right the first time around. It takes practice and a lot of trial and error, which can leave you feeling like you’ve been stuck on a paper for ages. That being said, writing is a chance to enable ideas to grow and change over time as part of the creative process, which can bring depth to your arguments and your story. You won’t get a perfect paper the first time around, so envision your time spent writing as constructive practice towards future perfection (or at least publishable perfection!).
Comments are closed.