How to Normalize Concentrations in a Fraction of the Time

Concentration normalization is boring. There, I said it. And I should know: I’ve had to do it more times than I care to remember. In fact, it’s probably one of the most tedious aspects of labwork. But did you know there’s a way you can do it in a fraction of the time? It’s true, and we thought we’d show you how in the best way possible: a race.

Want the TL;DR? My colleague Daniel Yip and I faced off in the lab with his manual pipetting against me and the Synthace platform. I swept the floor with him. (Sorry, Daniel.)

Want the longer version? Stick around! We’re going to cover:

• What concentration normalization is and why it’s so hard
• How to automate normalization with the right software and hardware
• What happened during our ‘normalization race’ (including a video of our efforts)
• What all of this means for the future of lab work

Let’s get into it!

What is Concentration Normalization?

The goal is simple: take a list of samples and make them all the same concentration or volume. In other words, concentration normalization is where you dilute an entire set of samples so that they're the same concentration, typically before using them for something else later on.

This is often a prerequisite for getting reliable high-quality data. Common applications include steps in next-generation sequencing, cDNA synthesis, qPCR/dPCR, or various protein assays.

When you work in a lab, it’s a ‘bread and butter’ protocol—one you’ll run fairly often. But it’s also one of the harder, more tedious things to get right 100% of the time. And that’s a problem.

Why is Concentration Normalization so Hard?

Any scientist dreads normalizing samples, especially in large numbers. Here’s why:

• Every sample needs a different pipetting volume, down to 0.01 uL, which can be nerve-wracking and tiring to get right each and every time
• Dealing with samples in plates makes it really hard (if not impossible!) to check which samples have already been normalized
• It takes a long time and needs constant focus, which is difficult in busy, distraction-filled lab environments

All things considered, it’s an error-prone, cumbersome, and tiring piece of work. The worst part? If you get something wrong, you likely won’t know about it until you get further on in your work. And that can only mean one thing: starting from scratch.

How to Automate the Concentration Normalization Process

What’s better than normalizing samples yourself? Getting a robot to do it. Trouble is, writing the code you need to do this can be difficult, time-consuming, or just not something you’re able to do. Isn’t it better to spend more time working on the science itself, anyway? We think so. Our secret weapon here is, of course, the Synthace platform.

Making the above workflow takes only a few minutes. You can simulate these steps in advance but, when you hit ‘schedule’ to run it in your lab, this is when the magic really happens. Here’s what the Synthace platform does:

• Automatically calculates your required dilutions
• Plans optimized pipetting actions on your robot
• Writes the machine code needed to instruct your robot (a Tecan Evo, in my case)
• Executes the workflow while you go off and work on something else

The burden of calculating and programming the robot disappears. The tiring, error-prone, soul-crushing job of manually pipetting potentially hundreds of unique volumes is given to a robot that won’t get distracted by someone asking if they want to grab lunch. With a fraction of the hands-on time as someone doing the same thing manually, your samples are safeguarded and ready to use downstream.

Head to Head: My Robot Versus Daniel’s Pipetting Skills

Our challenge, should we choose to accept it: normalize a plate containing 96 samples to the same concentration. I was the lucky one who got to use the Synthace platform and one of our robots. Daniel, though, was the human control. Sorry, Daniel. (Again.)

Step 1: Designing the Work

The clock starts! I open Synthace and Daniel opens a new spreadsheet. I upload the plate information and design my normalization protocol. Daniel inputs his formula and makes his calculations. Once he’s done, he prints out the list of 192 different volumes he’s about to pipette to 96 different wells. 

We both leave our desks and walk slightly faster than usual toward the lab…

Step 2: Getting Into the Lab

I head for our Tecan Evo where my protocol is waiting to run. All I have to do is put my sample plate in the right position, add a second plate with diluent, and make sure the tips are stocked. Daniel finds the first number on his list, sets his pipette, and starts. He makes one transfer, crosses it off the list, then moves to the next. He’s about to do this 191 more times.

Step 3: Pipetting

Daniel pipettes.

I hit ‘start’ on the Synthace software and leave the lab. It’s fine, the robot will do it for me and I could really use some coffee.

Daniel pipettes.

Twenty-one minutes later the robot is done so I come back and move my plate to the plate reader to quantify how well my samples were normalized by Synthace and the Tecan Evo. This plate reading step is also controlled by Synthace, so the acquisition protocol is set and the results are automatically uploaded to the cloud. 

Daniel pipettes.

Step 4: Reviewing the Data

Five minutes later I’m back at my desk, Synthace is open and my data is there. Everything looks good. Happy days!

Daniel pipettes.

He finally crosses the last number off his list. His neck hurts and he’s not sure about a couple of those transfers: did I add 54 ul or 45 ul? Did I hit well D7 twice? He moves over to the plate reader, selects the correct protocol, runs it, then waits as he emails the results to himself so he can check them on his laptop. He could use a break.

Speaking of results…

Synthace on the left, manual pipetting on the right

How to Normalize Concentrations with Minimal Fuss

Although this was a fun way to prove a point, there’s a lot at stake here. Being able to work like this has a lot of knock-on effects, and it’s not limited to normalization. This way of working applies to all lab work:

• Simpler to do high throughput work
• Increased complexity becomes easier
• Much less time spent on tedious lab work
• More time spent doing everything else you’re interested in

These kinds of changes compound over time. More complex runs, more runs in a day, a week, a month, a year… more time to think about the next challenge, more chance of moving faster to your next breakthrough… it all adds up. If you could work like this with everything you do in your lab, what would it mean for you?