Dr. Karmella Haynes

The first scholarly public product I help my trainees to produce is a review article. In essence, a review explains past work in a way that shows how they converge on a major insight. A great review is more than a description of papers on the same topic, it’s meta-research, similar to the the type of homework that is done to develop a solid research project.

This is why writing your first review article is a big step. It’s your chance to synthesize knowledge in a way that shows you understand the field, and more importantly, that you can help shape it. Whether you’re a PhD student hoping to get your name out there or a postdoc defining a niche, here’s how to do it right.

1. Start With a Real Idea, Not Just a Category

The best review articles are not just lists. They’re arguments. They make a case for an emerging opportunity, unsolved mystery, or innovative framework.

DO: Pick a central idea that’s exciting, timely, and rooted in recent science.
Example of a strong idea: “The untapped potential of toggle switches in immune cells.” This idea sets the stage for exploring synthetic circuits and real therapeutic relevance.

DON’T: Pick an idea that is just a catalog of things.
Example of a weak idea: “A catalogue of engineered receptors for CAR-T cells.” This may have educational value, but lacks a compelling thesis or new perspective.

2. Find Recent Papers That Support Your Central Idea

Once you’ve chosen your idea, build a list of recent papers (past 5 years, ideally) that support it. You want to highlight how the concept is technically feasible and scientifically important.

DO: Choose papers that directly support your central thesis.

DON’T: Pad your reference list with marginally related publications to make it look longer. This is a common mistake. If the paper doesn’t directly support your idea, it will only confuse reviewers and readers, and waste your time. Importantly, do not include any review articles at this step. More on that later*.

3. Annotate Your References Like a Pro

Make a spreadsheet or simple document with your references, and add a short explanation next to each one. This step will help you filter out low-value references and strengthen the logic of your review. For example:

GOOD: “Bistable toggle switch held an on-state for 48 hrs, a relevant time frame for immune cell activity.” (Strong connection)

USELESS: “Toggle switch was made in HEK293 cells.” (Weak connection)

4. Identify Subsections From the Literature

Once you have 10–20 well-annotated references, patterns will start to emerge. Group related concepts into 2–4 key subsections. For example:

1. Toggle switches compatible with immune cell dynamics
2. Clinically relevant input/output signals for immune control
3. Deployment strategies for gene circuits in ex vivo/ in vivo contexts

Each section should feel like a chapter in your argument, not just a bucket of similar papers. Now you are ready to write!

5. Explain Why the Data Matters

For each reference you cite, don’t just say what the authors did, explain what it means for your idea.

WEAK: “A toggle switch was built in HeLa cells {citation 1}.” If you can’t say anything beyond “They did X in Y cells,” the paper probably isn’t relevant enough.

STRONG: “The on-state was sustained for 72 hrs in HeLa cells, suggesting the topology can withstand perturbations from rapid cell division, a relevant feature in the context of proliferating immune cells {citation 1}.”

6. Incorporate Additional Relevant Primary Research References

As you write the main body of the review, you will start to notice technical statements that need a reference to support them. This is where you will start adding more supporting citations. In this example, citation 1 is the paper that reported the toggle switch, and citation 2 is a study that measured differences in proliferation rates of different cell types:

“The on-state was sustained for 72 hrs in HeLa cells {citation 1}, suggesting the topology can withstand perturbations from rapid cell division, a relevant feature in the context of proliferating immune cells {citation 2}.”

7. Efficiency is Evidence of Intelligence

When multiple papers support a similar technical point, distill them into a single, well-crafted sentence. This not only makes your writing easier to follow, but also shows intellectual maturity and control of the literature. Strive for both accuracy and efficiency:

WEAK AND WORDY: “A two-state genetic toggle was tested in HEK cells and held a weak on-state for two weeks {citation 1}. In Jurkat cells, a similar toggle showed half-the maximum level of output but the on-state lasted for three weeks {citation 2}. Another group tested the same toggle, but changed the output to GFP in a mouse model; while the signal diminished a bit over time, it was also stable, lasting for one month {citation 3}.”

STRONG AND ACCURATE: “Two-state toggle switches built with a double repressor topology show long term stability, albeit with a limited dymanic range across in vitro (HEK, Jurkat) and in vivo (mouse) models {citation1, citation 2, citation 3}.

8. Cite Other Review Articles – Strategically

*Pop Quiz – Remember the advice from Tip #2? No? That’s okay, here it is again: “Importantly, do not include any review articles at this step.” After you start writing, you can now start citing reviews with the following rules in mind:

Use reviews to compress well-trodden ground.
When your topic intersects with a major area that is not your focus, but still needs to be mentioned, summarize it in one or two sentences and cite a small number (ideally 1–2) of the most authoritative and up-to-date reviews. This shows your awareness of the broader context without dragging your article off-topic.

Never cite a review for a specific technical claim.
If you want to make a precise point, e.g., about circuit performance in a cell line, mutation effects, or dynamic ranges, go straight to the primary literature. Review articles often generalize or paraphrase. If you rely on them for details, you risk repeating oversimplified or even inaccurate statements.

Don’t take review articles at face value.
Skimming a title and abstract is not enough. If you’re citing a review to support a claim, at minimum, read the relevant subsection. Make sure the review’s actual content backs up your statement, and that it does so clearly and directly.

9. Save the Figures for Last – and Make Them Effective

Do not start with the figures. Start with your argument. Once your draft is structured, ask: “What parts of this are too abstract to explain with just text?” These are the places that need a figure.

Tip: If you would normally use your hands to describe it (or need a whiteboard), it needs a figure.

Avoid using fancy 3D shapes or drop-shadows to make simple figures look “cool.” If a figure looks boring, ask yourself whether it needs to exist at all. Make sure each figure communicates a real concept that words can’t.

10. Craft a Title That Serves the Reader

B now you probably have a provisional title. Think of your final, polished title as a software user interface or a readme document: it should help readers know exactly what they’re going to get. Avoid vague or flashy titles. Use your title to clearly signal what the review covers and why it’s important.
Example: “Genetic Toggle Switches for Immune Cell Engineering: Design Principles and Therapeutic Opportunities”

The best titles are complete, informative, and succinct, like a perfect tweet (in the early days of Twitter).

Final Thoughts

Writing your first review article is a chance to stake out intellectual territory. If you approach it like a mini-research project, anchored in a bold but feasible idea, organized around clear subsections, and framed with thoughtful context, you’ll produce something that’s both useful and memorable and is likely to be cited by other writers.

Need help turning your annotated references and draft into a compelling manuscript? You can book a coaching session with me here. I’m especially excited to support early-career researchers with interdisciplinary ideas and big-picture thinking.

Keep building. Your first review article could be the one that shapes the next wave of your field!