Wet Lab

A wet lab is a laboratory where work is done with real substances. There you find refrigerators full of cell cultures, test tubes, pipettes, and devices that spin or heat samples. The name comes from the fact that almost everything happens in liquid form: solutions, nutrient media, blood samples. Its counterpart is the dry lab, a room full of computers where nobody touches anything except the keyboard. There, data is analyzed and processes are simulated on screen. The term comes up frequently in news about artificial intelligence these days, because companies eventually have to verify their computer-generated predictions in a real laboratory.

The reality check for predictions from the computer

Software today can predict what shape a protein will take or which molecule might help against a disease. But such predictions are initially only well-founded guesses. Whether an active compound actually works and is not toxic only becomes clear with real cells. That is exactly what a wet lab delivers. It is the point where a calculation becomes true or false.

For companies, this step is the most expensive one. A single lab experiment can take weeks and cost a lot of money. That’s why it’s valuable when a computer model reduces the number of necessary experiments. Instead of testing ten thousand substances, you might then only test fifty promising ones. The savings lie not in the lab itself, but in the experiments you no longer have to run.

Investors now pay close attention to whether a biotech company can present its own wet lab results. Pure software promises are considered weak evidence. A common misconception is that AI could eventually replace the lab entirely. So far, it only shifts how much work has to be done in the lab.

From pipetting to biosafety levels

The basic work in a wet lab is surprisingly hands-on and manual. A pipette is used to measure out tiny amounts of liquid, often less than a drop. These samples go into plates with 96 or 384 small depressions, so-called wells. This allows many batches to be treated and compared simultaneously. Measuring devices then read out in which wells something has happened, such as a color change.

For results to be reliable, everything must proceed cleanly and be documented. There are biosafety levels ranging from S1 to S4, depending on how dangerous the organisms are. A level S1 lab works with harmless bacteria, while S4 works with highly dangerous viruses behind airlocks and protective suits. Every experiment also needs control samples, which show whether the procedure worked at all.

Increasingly, robots are taking over the pipetting. Such automated labs are sometimes called self-driving labs. In this setup, a program plans the next experiment, a robotic arm carries it out, and the measurement data flows directly back into the model. What was once a one-time test becomes a cycle of prediction and measurement.

Where the term shows up in the news

You most often read it in connection with biotech and pharmaceutical companies searching for drugs using AI. When a report says a candidate has been confirmed in the wet lab, that is a much stronger claim than a mere simulation. Research institutions also distinguish their job postings accordingly: wet lab positions are aimed at lab personnel, dry lab positions at programmers and data analysts.

A second area is real estate and location policy. Lab space requires fume hoods, gas lines, and waste disposal — a normal office cannot simply be repurposed. In cities like Boston or Munich, wet lab space is therefore considered a scarce commodity with high rents. Reports about new lab campuses regularly appear in the business section.

Finally, you may encounter the term in school without it being called that. The biology or chemistry classroom is basically a very simple wet lab. There too, the rule applies: measurements are taken on real material, and a result only counts once the experiment is repeatable.