Scientists often use experimental kits that come with prepackaged enzymes, buffers and substrates for biomedical research.Credit: Hugh Threlfall/Alamy
Since my postdoctoral years at Louisiana State University in Baton Rouge more than three decades ago, I have witnessed a profound change in how biomedical research is conducted. This change has made research much more productive, has strengthened experimental protocols and has helped to ensure that findings are replicable. But it has also led to a huge increase in costs and, for some scientists, perhaps a loss of fundamental understanding about experimentation. The age of the biomedical research kit is upon us.
Kits used for biomedical research are commercially produced, pre-packaged experimental systems designed to perform a specific assay or measurement with minimal preparation by the user. A typical kit arrives as a small box containing pre-measured reagents — often enzymes, buffers and substrates — aliquoted into tubes or plates, alongside a concise protocol. The assays are designed to be straightforward, reducing the amount of preparation and optimization performed by the user. In minutes to hours, most protocols yield a numerical signal, such as a measurement of absorbance, fluorescence or luminescence. This output is then interpreted as a representation of a biological phenomenon of interest.
Experimental kits span a wide price and complexity spectrum. Simple assays might cost up to a few hundred US dollars. Meanwhile, assays designed to investigate specific cellular pathways, kits for preparing DNA libraries, panels that can measure several factors simultaneously and kits required when working with specific instruments all frequently range from several hundred to many thousands of dollars.
Useful tools
In my experimental work, we often use kits to quantify mitochondrial function, oxidative stress and cell viability and cell-death mechanisms, all techniques that once required the preparation and validation of various reagents in-house. These kits commonly cost several hundred dollars for a limited number of assays.
AI linked to explosion of low-quality biomedical research papers
Researchers use experimental kits to generate large volumes of data, yet the underlying chemistry, biological principles and assumptions that the assay makes can easily fade from view. Using a kit without understanding its principles is like relying on a calculator without grasping the underlying arithmetic: the tool functions smoothly, but interpretation and troubleshooting become more difficult when results depart from expectations.
One could argue that modern science depends on abstraction. People do not need to understand how a clock works to tell the time. Similarly, many researchers might not need to know the precise composition of every proprietary buffer to answer meaningful biological questions. The field of computer programming is having a similar conversation because artificial-intelligence tools are increasingly turning once-essential skills into optional ones.
The challenge, then, is not deciding whether these research ‘black boxes’ should exist — they will — but finding out which skills are essential for scientific judgement, troubleshooting and integrity.
Changing times
The problem arises when assay kits become a substitute for understanding. When troubleshooting is reduced to swapping brands or ordering a fresh batch of reagents, laboratory staff lose an essential skill: the ability to interpret failure as information, not merely as an inconvenience.
When I was a postdoc, a lot of emphasis was placed on ensuring that students understood the science underlying their research and had basic lab skills. We didn’t have kits to challenge that. One of my friends recalls waiting outside his lab for his professor. On arrival, the professor asked him several simple questions: one about a specific equation crucial to their research, one on molarity calculations, one on preparing a buffer and one about titration. He was given the keys to the lab only after answering. Another colleague told me that his professor deliberately kept the lab’s kymograph (a tool for plotting data) dismantled, just to see whether students could reassemble and use it correctly.
