From a one-line description to a graphical abstract: a CRISPR-Cas9 knockout walkthrough
In my last post, How I model scientific claims before generating a diagram , I wrote about the thinking that happens before any image exists. This one is the hands-on part. I'll take a short text description of a CRISPR-Cas9 gene knockout, turn it into a graphical abstract, and then go through the output the way I'd review a figure from a co-author. I built Scientific Figure Generator, so I'm…
In my latest post, titled "How I model scientific claims before generating a diagram," I demonstrated the process of transforming a succinct text description of a CRISPR-Cas9 gene knockout into a graphical abstract. Using my tool, Scientific Figure Generator, I followed a five-step knockout workflow, which included guide RNA design, Cas9 delivery into cells, double-strand break, NHEJ repair causing an indel, and loss of protein expression.
To create the prompt, I typed: "Graphical abstract: CRISPR-Cas9 gene knockout workflow — guide RNA design, Cas9 delivery into cells, double-strand break, NHEJ repair causing indel, loss of protein expression. Flat illustration style, clean labels." This brief, structured prompt allowed the model to focus on the key stages of the knockout process without unnecessary details.
By naming the figure type, listing the stages in order, and specifying a flat style, the tool was guided to produce a clean, thumbnail-sized illustration.
Upon generating the image, I reviewed the output using a review checklist. Scientific accuracy was assessed panel by panel, ensuring that each representation was appropriate for a general audience while still offering enough detail for reviewers in the field. For example, the guide RNA design panel depicted a guide RNA with a stem-loop paired against a DNA helix, which, while not strictly accurate, served as an effective icon for the step.
Similarly, the Cas9 delivery panel showed the enzyme docked at the edge of a cell, simplifying the complex process of delivering the enzyme into the nucleus.
When checking label correctness, I verified each label letter by letter to minimize the risk of garbled text. The final output featured five panels, each labeled appropriately and depicting the correct sequence of events in the knockout process. All labels matched my original description, and no errors were found in the arrows or overall logic of the figure. For publication, I would choose a higher resolution (2K) and adjust the style as needed.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
