How to go about writing a paper.

When writing a paper, the decision of where to start is up to each person. Some people like to start by writing the results right away, others like to focus on the introduction first, and some like to start with the methods. Start with what feels easiest to you and see if you need a change of strategy only when you hit a roadblock.

If you don’t know where to start, then as a general rule it is suggested to start with the results. Anecdotal evidence suggests that people find this section easiest to write. Also, after the results are written, they act as a compass against which all the other parts are based.

Things to consider:

  • Abstract: How can I concisely communicate the essence and significance of my results?
  • Introduction: What information does the reader need in order to make sense of the results?
  • Discussion: What new insights or questions can I bring to the attention of the reader now that she knows the results?
  • Methods: What methods were required to obtain my results?

What is the main message or conclusion of your paper? While there are many effective writing styles, organization and structure are essential to any manuscript. Here are some ways you can organize your thoughts:

  1. Look at all the data and subdivide it into experiments you’ve performed.
  2. Write down the most important conclusion drawn from each experiment.
  3. Determine how these conclusions relate to each other and how can they be ordered so as to create a logical path. (Often this will not be the same order in which you performed the experiments.)
  4. Make an outline that presents the experiments in this order. How does this arrangement affect the structure of the introduction? The punchline of the abstract? The ground-breaking question set forth in the Discussion?
  5. If you do not feel convinced, try rearranging your experiment order and start again. Sometimes multiple iterations of this process can reveal that an additional experiment should be included or that certain results are not necessary to include in the main text.

If you are still struggling to organize your thoughts: Try giving a short oral description to a colleague or friend. It doesn’t take much time and sometimes you’ll realize that you intuitively know the logical flow of the paper.

  • A colleague who is somewhat familiar with your work can help identify the logical path of the experiments
  • Your non-scientist friend or relative can help you figure out how to clearly describe the main points of the paper

If these approaches don’t help, just start writing. Even with an incomplete outline, sometimes just getting words down on paper can bring clarity to the process.

Tips on organization

Any person who has given a presentation or written a paper or grant proposal knows the agonizing frustration of pouring through your lab notebook trying to locate data from experiments performed months (or years) prior. In retrospect, one realizes, it would have been nice to have kept your data in a ready-to-go format so that when it comes time to put everything together for formal presentation, you can save yourself some serious time. Here are a few ideas on how to decrease the stress of compiling and presenting your data:

Keep your data figure-ready

  • Record experimental details. This sounds obvious, but it can be easy sometimes to forget to write down concentration units, or graph variables. After time has passed, this information may not seem as obvious as when you originally performed the experiment.
  • Perform “publication-ready” experiments in a timely manner. As soon as you obtain the results of a “successful” experiment, verify that all the necessary replicates, appropriate controls and statistical tests (if relevant) have been performed. Going back months later to re-do these experiments with better controls or more replicates can be very frustrating!
  • Format your results into figures. Keeping all your data in “figure format” will expedite the process of writing, and will ensure that you are prepared in the event that you have to make a last minute presentation.

Maintain organization in your reference library:

  • Consider using an organizational tool. Applications such as Papers or ReadCube are great options!
  • Keep track of methods references. If you used a technique that you will need to reference, write this reference next to the experiment. Finding these papers again months/years later can be time-consuming.
  • Keep a running list of references that could contribute to an Introduction or Discussion section of a manuscript.

Always ask for feedback as your data evolves:

  • Show colleagues your figures. Not only can they give you feedback on your experiments, they can let you know if your data is presented in a clear manner.
  • Ask for suggestions for better or additional experiments. While some labs have group meetings and subgroup meetings that allow them to get constant feedback, others do not. Maximizing feedback prior to submission helps ensure that you are submitting an optimized product and can help you to anticipate what reviewers may request after submission.

Before Writing

The decision of where to publish is an essential part of writing a paper. This decision greatly impacts how the paper will be written (for what audience), what experiments are necessary to include, and the overall format (length, figure number, reference format) of the manuscript. Here are a few things to consider before starting to write up a manuscript.

Most journals have a section on their website called “Author Guidelines.” This is where you can find any relevant information needed to submit to this journal. Once you have located this information, make a checklist of items required by the journal.

For example:

  • Is a cover or submission letter needed?
  • Does the journal provide a manuscript template? (This can greatly expedite the formatting process)
  • What is the required citation format?
  • Are only certain types of files accepted? (many journals specify preferred or required file types and sizes)
  • Are specific fonts required for figures and text?
  • Is a graphical abstract requested?

Having a clear sense of what the journal requires before writing up your manuscript will minimize the headache of going back last minute to re-format, or change figure fonts or file sizes. While seemingly minimal, these changes can accumulate and cause unnecessary last-minute stress.

Avoiding Jargon

Avoiding technical nomenclature can make your work accessible to a more diverse scientific audience, but at the same time you don’t want your abstract to be so high-level that an expert reader is bored and left wondering what, specifically, you did. As a compromise, provide a general description of any jargon-y words and then include the more specialized name/detail (ex. rather than “16S rRNA”, you might say “the bacterial taxonomic marker gene, 16S rRNA”). A good way to catch jargon is to have a colleague or friend in a different area of science read over your abstract and point out any terms that obscure your meaning.

Adapting your writing style for expert vs novice audiences

Expert. A small group of scientists in your niche sub-field, or “experts”, will have prior understanding of 1) your area’s state-of-knowledge, 2) the major questions that need answering, and 3) the commonly used experimental techniques and analyses. You are most likely to encounter such an audience when publishing in a journal that focuses on a specific area. To adapt your writing for expert readers:

  • Cut to the chase. Shorten the general background, then get even more specific in your “specific background”.
  • Include more nuance. You can afford to include more detail in your motivating question and methods since you will spend less space catching readers up on definitions, etc.

Novice. Often, “novice” readers, or scientists that are are not in your specific subfield, will comprise a much larger percentage of your audience. By making your work accessible to novice readers, you can significantly increase the number of people that your paper reaches. In this case, the introduction will play an extremely important part, making the motivation for your work and your findings accessible to novice readers. To adapt your writing for novice readers:

  • Provide more background. Take a few steps back in your general and specific background. You will sacrifice detail for clarity; you may not narrow down to as specific of a “specific background”, and you might expand the entire background section.
  • Avoid jargon. Introduce/define all the “key players” and important techniques that will be addressed in the paper.
  • Simplify the motivating question and results. The more basic the question posed in the Knowledge Gap, the more people will care about it. Take the time to identify the simplest, most fundamental question your work answers.

This content was adapted from from an article originally created by the MIT Biological Engineering Communication Lab.