How to Write a Scientific Review Article

By Leila Haery

Writing on a notepadWriting a review article is a wonderful way to develop and exercise your scientist skill set. If you dread the thought of writing a review, or if you’re currently stuck trying to write one, hopefully this post will help you get things moving - remember you're becoming an expert in your field and are the perfect person to be writing the review! Doing so is a great way to develop your ability to write, to read efficiently, to search the literature, and to synthesize a large volume of information: basically, a scientist’s tool kit.

Click here to contribute to the Addgene BlogI wrote a review article on histone modifers in B and T cells with my adviser when I was a graduate student. Naturally, the first thing I did was google “how to write a scientific review.” The second thing I did was stare in horror at the limited number of hits. So, here, I will share my process with you. But before I do that, let me say that I was by no means an expert on this topic when I set out to write the review. That’s just to say that you don’t have to be one either.

Choose the topic and outline the organization of the review

Once you start reading, there will be a temptation to include every piece of information that was ever published. Obviously this isn’t possible. So, define your scope from the onset. Perhaps you, a colleague, or your adviser was invited to write on a particular topic. Alternatively, maybe you’re researching a topic for which no relevant or recent review exists. Once you pick a topic, try to be specific about exactly what aspect of the field you plan to review. If it’s a well-researched field, you may need to get specific to make sure your article doesn’t turn into a textbook.

Get the journal’s submission rules for review articles

Whether submitting a review by invitation or by your own accord,  once you have these rules (word limit, formatting guidelines, etc.) you have some criteria to shape the document.

Get and use a reference management program (e.g., EndNote, Papers, Mendeley, etc.)

You’re going to be managing a lot of references. I cite as I write, meaning I use the software to add the citations in real time as I write. Things are going to get a little crazy (meaning you are probably going to cite hundreds of references) and it’s better to keep your references organized from the beginning. I also recommend using the citation style of (Last name, Year) in the document while writing, because it helps you later on to remember where you read particular studies or experiments. Later, you can easily convert the citation style to whatever the journal requires. Using the (Last name, Year) format also has the benefit of exposing you to relevant researchers in the field. Finally, you can sound credible and cool when you casually mention “Haery et al., showed that MYC expression was increased…” when discussing the review topic with your peers.

Review Article FigureStart reading!

I started by reading other reviews because, as I mentioned, I wasn’t an expert in the field. To find reviews, I just searched online and found ones that I thought “looked good” by no definitive criteria. I read these articles to get a sense of the themes in the field and to learn what people cared about. I also used reviews to get a list of research papers that I needed to read. Once I had an idea of the themes in the field, I searched for recent research papers on these particular themes, for seminal papers on these themes, and also for articles from the active/well-known researchers in the field. I made sure to find information from genome-wide studies, as well as results from smaller and more specific studies. I also did not limit myself to the well-cited or popular papers, but looked for papers from a wide range of authors.

Just start writing

When I first started I thought I would read a bunch of papers and then feel ready to write. What actually happened was that each paper taught me a few things and also highlighted a few dozen things that I didn’t know about. Instead of reading a paper and getting my bearings, I would read a paper, panic, and then download a bunch of other papers. In mathematics, I think this is represented by factorials. In environmental science and ecology, this can be represented by the tip of the iceberg. For writers, this is probably “a normal day.” The way I broke this cycle was to just start writing.

exclamation.jpg__65x63_q85_subsampling-2_upscale.pngA word of advice: when reading, don’t feel like you need to read every paper from start to finish. Find the information that’s relevant to you, note it, and move on. At the beginning I felt like this was cheating. Eventually I learned that it’s practical because there simply isn’t enough time to read every single paper. I also learned to appreciate the discussion sections of the primary literature as a resource that both summarized results and contextualized them.

No really, just start writing!

Don’t worry about grammar or formatting or continuity. Also don’t worry if you feel like you don’t still know enough about the topic. Just get as many words down on the page as possible. This could be in the form of lists, streams of consciousness, or anything else. Like I mentioned, I also added citations in real time as I wrote, so each statement was referenced even in the roughest version of the draft and I didn’t have to worry about having to hunt down sources later on. 

exclamation.jpg__65x63_q85_subsampling-2_upscale.pngA comment about writing: initially I felt that I needed a huge chunk of time to sit down and get anything done, so unless I had at least a two hour block, I wouldn’t work on the review. This quickly became a problem when I realized how elusive a free two hour block of time was. Eventually, I learned that I was most productive in short bursts (even twenty minutes was enough to make some progress). For most researchers, it’s not always possible to pull a Thoreau and retreat to the woods for months at a time, so take advantage of whatever time you have. If you’re like me, you might be surprised at how much you can get done between office hours and experiments.

A comment about citing: I wasn’t sure if I should cite reviews or the primary literature or both. I ended up citing both because I used both.

Curate and present some useful data

Typically lists and pictures are the most useful parts of reviews. These could be in the form of figures/schematics or tables. And don’t forget to include citations so that people can go back and read the original reference for the data. For example, we summarized how frequently each member of a class of proteins was mutated, as reported in various studies. We made a table that listed each protein in the class, then for each protein we listed all the studies that reported mutations in that protein (including how frequently a mutation was found and the size of the study). This was useful because you could easily see how frequently each protein was mutated, you could see how big the studies were, and you could find the original paper if you wanted to learn more. Another example: we made a schematic of all the proteins in the class that showed the relative sizes and the conserved domains. This information was available in GenBank, but it was useful to present it all in one place to get a sense of the similarities and differences among proteins in the class.

Curated Table Large.png

Offer your perspective

It doesn’t have to be long and it doesn’t have to be revolutionary, but you could include a few comments on where you think the field is going or what areas are worth exploring. 

Edit and rewrite, then repeat

If you’re like me, your rough draft is really rough and so editing is going to be a long process. I was lucky in that my adviser always played an active role in writing and editing, so I always had someone to send drafts back and forth with. That being said, we probably exchanged dozens of drafts of the manuscript. This is the time to transform the document into something cohesive-- change the sentences, make it flow, and start telling the story. Like any editing process, you will need time away from the article to be able to keep editing it effectively. You will also begin to hate the article. This is normal and it means you’re on the right track! 

Get feedback

After you’re done editing, send it to some actual experts in the field for their feedback on the scope and the content.

Submit your article!

Overall, writing a review can be overwhelming and challenging. My best advice is don’t overthink it. And to quote my adviser, "you just gotta do it." At the end of the day, someone has to write the article and that someone is going to be you. So, just do it! I will also paraphrase what I have heard many other creative people say about writing: you don’t know what it’s going to look like when it’s done, but you know what it looks like when it's not done. So, as long as it doesn't look done, just keep working on it. Also bear in mind that this is just a review article and not your life’s work: so remember that done is better than perfect. Good luck!


1. Haery, Leila, Ryan C. Thompson, and Thomas D. Gilmore. "Histone acetyltransferases and histone deacetylases in B-and T-cell development, physiology and malignancy." Genes & cancer 6.5-6 (2015): 184. PubMed PMID: 26124919. PubMed Central PMCID: PMC4482241.

Additional Resources on the Addgene Blog

Resources on

Topics: Scientific Sharing, Scientific Publishing

Leave a Comment

Sharing science just got easier... Subscribe to our blog