How to write scientific papers

See this useful writing guide by Matteo Carandini.

• Tips on writing good paper titles.
• The shortest ever guide to academic writing.
• Writing in an accessible manner.
• To write well, read your writing out loud. Every first sentence of a paragraph should summarize it (see the guidelines by Matteo Carandini)

Practice version control; I like Google Docs to avoid conflicts and unclarity about versioning, and it works well with Zotero. Make sure to name important versions of the manuscript like ‘preprint_v1’. However, if you prefer another system, I’m happy to be convinced.

Before submitting a preprint/paper with me

Send all co-authors an email with the following:

• Final paper draft that has been approved by all co-authors
• Journal & preprint server this will be submitted to
• Reviewers to suggest
• Link to public data
• Link to public code. Has the code been reviewed?

Note: I must be able to reproduce all figures from your public data and code before submission.

I’m a sucker for nice-looking figures (and fonts).

Learn how to make your figures look good straight out of Python/Matlab/Julia/R, and minimize the time spent changing things in Illustrator/InkScape. This will not only save a lot of time in the long run, it also makes your analyses more reproducible.

Here are slides of a figure workshop we did in lab meeting on 26 February 2024. Watch back the recording when you’re starting to make your own figures!

Ideally, follow this IBL styleguide for consistent and good-looking figures.

Some other resources:

• Learn to use colors to your advantage: Borland D, Taylor II RM (2007) Rainbow color map (still) considered harmful. IEEE computer graphics and applications 27:14–17.
• Tufte ER (2001) The Visual Display of Quantitative Information. Graphics Press.
• Weissgerber TL, Milic NM, Winham SJ, Garovic VD (2015) Beyond Bar and Line Graphs: Time for a New Data Presentation Paradigm. PLOS Biology 13:e1002128.
• Ten Simple Rules for Better Figures

Authorship

It is unfortunate that we don’t talk about authorship (early) enough. While I have some opinions, we should discuss authorship for each project.

• Scientific autorship is a tricky business (see here for discussion in my PhD lab). I aim to discuss authorship early on, clarify expectations, and give regular updates when a project changes. If you are unsure about authorship, please let me know asap.
• I aim to have an honest authorship discussion at the moment a project looks like it might become a paper (this is not necessarily always the case when we work together, for instance when you do a short student internship).
• Early on, we will try to settle on an initial authorship order that everybody is happy with, and clarify the expectations of everyone in the group. When the work changes, new people come on board or someone leaves, we should discuss again.
• If you feel unsure about authorship, bring it up asap: I much prefer to have an early discussion and avoid later misunderstandings or frustration.
• Whatever authorship scheme we settle on, I encourage using the Credit scheme with additional detail to keep track of everybody’s contributions. Ideally, we add this statement to the preprint/paper together with a clear tabular visualization.
• Apart from traditional authorship order, we will indicate each person’s contributions to the project in a tabular form (inspired by Nick Steinmetz, further develop in IBL).