Context

Sooo, it finally happened! I recently landed a new job at Roche in Basel, Switzerland and this new role I’m assuming is that of a Lab Automation Engineer (LAE). This is extremely exciting and it finally lets me break free from pure fullstack development into a more challenging domain. Not to mention the fact that it’s a new country I’m moving to, that is roughly a thousand kilometers away from home, which just hypes me up even more. Needless to say, I’m counting the days until I get started (which is ~3 months as of writing this).

In university I was overly eager to start learning about my classes before the semester even began and it seems as if I still carry this property even after graduating 5 years later. Currently, the issue is that although I would say that I’m a pretty good generalist builder, I’m certainly lacking domain expertise in the pharma industry. But I still have 3 months to prepare, so better get started.

Naturally, one of the first questions one might have is: what the heck even is a lab automation engineer? Let’s find out.

What even is a lab??

Before we can answer what a LAE is (and does), we first have to look at what a lab even is. We all have an image in our heads: someone in a lab coat, holding a flask with some coloured liquid in it and a bunch of microscopes in the background. But perhaps we can find more grounded information. Let’s ask Google. A few search results later, we find the website of Fraunhofer and specifically this image:

Now that is a lab if I ever saw one. Scrolling down a bit, we can see a YouTube video with another interesting photo:

By eyeballing these two images - and currently being pretty oblivious about labs in general - I would say that the first image seems to be more automated than the second one. But that is not really the point right now. What is interesting is what kind of devices we can spot in these images and then ask ourselves, what other devices can we usually find in labs. Each of them has a special purpose, usually doing something to some liquid.

There is no shortage of resources on the internet on this topic and you could probably fill a whole book on this, but I only have 3 months time. The 80-20 rule must be used here. We can look at this Wikipedia page to get an overview but another good resource is any site that sells lab equipment (which also teaches you about what brands are out there).

As humans, we like to categorise and put things neatly in boxes and that’s what I will try to do here. I think in general you can split up lab equipment into two broad categories: (expensive as heck) electrical devices (think oven, liquid dispencers, etc.) and single-use (-ish), generic tools (like IDK a pair of scissors or a well). The former is most interesting to us, but we shouldn’t neglect the latter either. For now, we will focus on these electrical devices and if we have some time towards the end, also look at the most common instruments of the latter category.

Frequently Used Lab Equipment

(Disclaimer: I have never worked in a lab! This list is just what I gathered from my research on the internets!)

Ok, so a lot of the work in a lab is just moving liquids. Be it within the machine itself (e.g. shaking the liquid, spinning the liquid, etc.) or distributing the liquid into separate wells.

Pipettes

We start simple, with a relatively cheap, but probably very often used device: a pipette!

By “relatively cheap” I mean in relation to the other devices we will see later in this list. Each of these bad boys can set you back between 100€ and 350€! From what I can tell, the Eppendorf brand appears a lot and seems rather expensive; I can’t speak for the quality yet but so far they seem to be a bit like the “Apple” in lab equipment. Fun fact: Eppendorf is the name of a district in Hamburg and as a student I used to live there. Another fun fact: just like the lab equipment, Eppendorf is one of the more beautiful (and expensive) areas to live in Hamburg.

Shake It Up!

Shaking liquid is also common practice in a lab. I’m not 100% sure why, but the analogy I have in my head is that of juice (or oat milk, same idea). The juice particles sink towards the bottom and settle there (they are the suspended solids). Drinking unshaked juice tastes bland (you’re basically just drinking fruit-water without the fruit mixed in). After shaking it, the fruity-bits and the water have mixed and your juice tastes jucier. I’m not exactly sure what the right counterparts in a biotech lab are for this analogy, let me just quickly ask Dr. Claude…

Aaaand here is the response: The above section - while true - is not actually the main reason for shaking. What I described earlier is about homogeneity, but the main reason is for aeration. AFAICT, cells need to “breathe” oxygen. Air-y water only exists on the surface at the boundary between liquid and, well, the air. If you shake the liquid, the surface gets renewed (air-y liquid from the surface mixes into the water, non-air-y water moves to the surface, absorbs a bit of oxygen, moves away, and so on) providing the cells with oxygen there and not making them suffocation to death. Otherwise, growth of the cells would mostly be concentrated near the surface and thus be limited.

Here’s an image I drew that may (or may not) help your understanding:

My analogy above is also still valid: by shaking the liquid, instead of having one dense sludge of cells at the bottom, they are evenly distributed in the liquid such all cells kind of see the same particles (and nutritients) around them.

(Shaking is also used for speeding up reactions and dissolving/resuspending)

Let’s start with an orbital shaker:

This unit will set you back around 4000€. Yes. You saw that right. Here’s a video that shows you what it does. It literally just shakes the platform in a circular fashion. Why it costs so much is beyond me.

Another way to shake liquids is to use a vortexer. A vortexer kind of just wiggles its plate around and is often used for suspension (e.g. of cells). Think back to our juice and the juice particles that have settled at the bottom. If you were to wiggle the plate on top of which it stands, then the juice particles fly into the liquid and suspend “midliquid” and fly around there. Here’s one from the “Biologix” brand:

(Btw. nothing here is affiliated in ANY way, I just try to find out some of the common brand names as well)

You can also often find something called a shaking incubator. The shaking part we already covered, so what’s an incubator? Well it’s just an enclosed container that keeps a set of parameteres constant (mostly temperature). So a shaking incubator is an orbital shaker that ensures a constant temparature. Here’s one from Eppendorf:

They can also look a bit like a microwave:

The latter one sets you back around 23000€, equipment so expensive, you can start to use scientific notation ($2.3e4$).

Often times, you want to grow cells to test whatever drug or compound you’re developing. But cells don’t just grow under any condition. They need a specific environment which stays constant, e.g. with a constant temperature or humidity. E.g. if you’re growing human cells, you should make their environment close to our bodies, i.e. at a constant ~37°C.

Centrifuges

A centrifuge is a device that simply spins something and that something then experiences centrifugal forces. I’m sure most of us have been on a merry-go-round. On those, you can feel the centrifugal forces that push you out. Here’s an image:

If you have the misfortune to ride these death traps, you will have experienced it too. They say astronauts can withstand who-knows-how-many-gs but the final NASA test is to last in these for longer than 3 minutes.

When you have some proteins floating around in a liquid, then gravity is not enough to overcome the Brownian motion. You also don’t really inspect individual proteins: in biology everything is so small that it moves towards statistics instead. You want most of your proteins to be at the bottom and hopefully most of those are the proteins that you were actually looking for. You also can’t really “look” at a protein anyway (they are smaller than the spectrum of visible light). After you’ve moved most of the proteins towards the bottom of your container (which is what the centrifuges do), you can extract those, do some purification process and then perform mass spectrometry so you can be mostly certain about what protein you were working with.

Here is a (chilled) centrifuge from Beckman Coulter:

I couldn’t find the exact price for this (you have to request a quote), but it’s probably ~6000€.

Nuclear Bioreactor

We had already mentioned incubators. They control the environment of a closed off room, e.g. by holding a certain constant temperature. A bioreactor is related to that but instead of controlling the environment, it controls the culture. So there is a bunch of cells in your reactor and you want to control their “state”. To do that, you have a few probes that read out metrics like pH, oxygen, nutrients, etc. This is something you might want to do in order to e.g. force the cells to produce a certain protein, something they might only do under the right circumstances. The bioreactor is basically what achieves and maintains that state.

From the software point of view, a bioreactor is a constant stream of data that you have to log, understand and finally react to.

These can get very expensive and for the top tier ones, scientific notation starts to make sense unironically. The one above costs around 18.000€, which is on the cheaper side from what I could find.