Laboratory Automation – Robots for Life Scientists

April 19. 2018. 6 mins read
Table of contents

When we first read about the incredible potential of nanotechnology as described in Engines of Creation, the only thing we could think of was how to invest in it. The potential for molecular machines or “nanobots” that could be programmed to do just about anything sounded limitless, and we wanted in on the ground floor as investors. This was back in 2003 when G Dubya signed the Nanotechnology Research and Development Act, but since then things seem to have failed miserably. Carbon nanotubes didn’t do jack, although Nantero is still kicking. Other nanomaterials like graphene have resulted in some expensive sporting equipment and little upside for investors so far. More noticeably, the prefix “nano” and the word “nanotechnology” have all but disappeared, but the ideas haven’t.

Scientists are now using technologies like gene editing and laboratory automation to build synthetic organisms that do cool things like “sweat” biofuel. In fact, there are now companies like Ginkgo Bioworks that use lab automation to produce thousands of new prototype organisms for their clients. We first visited this topic back in 2016 when we wrote about “3 Companies Building Nanorobot Factories“, and we talked about how these new synthetic organisms sound very much like the nanobots we’ve been waiting for. More recently, we revisited the topic and pointed out that Ginkgo Bioworks was likely to reach a $1 billion valuation. Indeed, they achieved unicorn status just days later – an event that was probably equal parts encouraging and discouraging for the four other players in this space.

The automated production of organisms that serve humans is made possible by something called laboratory automation or lab automation, and companies like Ginkgo Bioworks and Zymergen use it in-house to build alien organisms. That’s one application of lab automation we’ll call “in-house”. Then there’s something better described as “laboratory automaton in the cloud”, and we first talked about it a few years back in our article on Emerald Therapeutics: A Robotic Laboratory in the Cloud. This is where anyone can run experiments on expensive laboratory equipment that lives in “the cloud“. This type of lab automation makes an entire laboratory available to aspiring life scientists who don’t have the capital needed to purchase expensive life sciences equipment.

So what’s been happening with Emerald Therapeutics since we last looked? Not a whole lot. They changed their name to Emerald Cloud Lab (ECL) and deployed a nice new website which provides a slick walkthrough of how their technology works, alongside a roadmap for future laboratory offerings:

It looks like the company is expanding and hiring, but they are being very tight-lipped about their progress, which could mean some sort of liquidity event is nigh. Or they could be playing their cards close to their chest, since there’s another company out there offering “a robotic cloud laboratory for the life sciences”.


Click for company websiteFounded in 2012, San Francisco startup Transcriptic has taken in $27.8 million so far to develop what looks to be the exact same sort of offering as Emerald Therapeutics. (Companies always love it when you say they’re just like their competitors). One similarity between the two startups is that they both share some of the same investors, like Peter Thiel. Then there’s venture capital firm Data Collective, who liked the idea of lab automation so much that they were lead investor in both laboratory automation startups and the aforementioned Ginkgo Bioworks – which may tell us something.

Transcriptic was in the news this past week when Ginkgo Bioworks and Transcriptic received a $9.5 million grant from DARPA so that they could throw some machine learning into the mix.

That isn’t the first time these two companies will be working together. Back in October of last year, Ginkgo Bioworks and Transcriptic entered into a five-year agreement valued at more than $10 million, which involved Ginkgo incorporating Transcriptic’s cutting-edge robotic automation software into its Boston-based foundries. Here’s a blurb from the press release announcing the partnership:

“Transcriptic’s ability to translate organism designers’ vision into reality via lab automation is unparalleled, and brings an unprecedented scale to our organism foundry,” said Barry Canton, Ginkgo Bioworks co-founder. “

Remember that Ginkgo Bioworks is a $1 billion company that’s taken in a whopping $429 million in funding. The fact that they made that statement about Transcriptic is a pretty solid vote of confidence. Does this mean that Transcriptic might be moving towards marketing their tech to “in-house” lab automation applications, while Emerald Labs will focus on  “lab automation in the cloud”? Who knows, but Transcriptic is being much more open about their company happenings, like the 2017 Year in Review update they published on their blog which gives us some interesting metrics. For example, 79 million micro-liters of fluid were pipetted by their robots in 2017, making us wonder what exactly a “pipette” is.

Now we’re not experts on what happens in laboratories, aside from one of our MBAs who unsuccessfully tried to make LSD once in chemistry class. For that reason, we want to look at an easy-to-understand example of lab automation, and a pipette sounds like a good place to start. For that, we’ll turn to a startup called Opentrons.


Click for company websiteFounded in 2010, Brooklyn New Yaaaawk startup Opentrons has taken in $10 million in declared funding so far, nearly all of which came in the form of a round led by Khosla Ventures which closed late last month. That money is being used to build an automation system called the OT-2, with the earlier model OT-One being “used by 90% of top 50 research universities and 7 of the top 10 pharmaceutical companies”. Here’s the first sentence in the “about us” section:

Today, biologists spend too much time pipetting by hand. We think biologists should have robots to do pipetting for them.

We immediately turn to Wikipedia where we learn that “A pipette (sometimes spelled pipet) is a laboratory tool commonly used in chemistry, biology, and medicine to transport a measured volume of liquid”. Here’s what a basic pipette looks like:

A couple of pipettes

You can get a pack of 100 pipettes on Amazon today for about 6 bucks. If you’ve ever used an eyedropper, then you know how a pipette works already. You use your thumb and forefinger to squeeze a plastic bulb so that fluid is either pulled into or released from the pipette. This process is then known as “pipetting”. Now imagine that you had to fill a 96-well plate so that each well contained an exact amount of some chemical. Here’s what a 96-well plate looks like:

Imaging filling each of these 96 holes with various types of fluid

How long do you think that would take you? Better question, how boring would that job be? That’s why Opentrons built a robot that can “pipette” a 96-well plate in 20 seconds. No more worrying about manual mistakes or inaccurate measurements of fluid. “Opentrons electronic pipettes take the classic micropipette design and simply replace the scientists’ thumbs with electric motors” says the company, which frees up your highly paid scientists to add more value doing science stuff. The company’s pipetting robots come in various flavors as seen below:

The robots range in price from $3-4 thousand, and a single electronic pipette will run you about $500. If you think that’s expensive, it’s not. Competing tools run 10X that much, which is why 90% of the 4 million biologists around the world still do their “pipetting” by hand.


The pipette robots from Opentrons are an example of just one type of lab automation that is helping us move towards fully automated laboratories that can be accessed by anyone with a credit card. Lab automation promises to democratize access to expensive equipment, which should increase innovation. For existing laboratories, this means doing things faster, better, and cheaper. According to the U.S. Department of Labor, in 2016 there were 335,700 medical laboratory technologists/technicians involved in “collecting samples and performing tests to analyze body fluids, tissue, and other substances”. With a median pay of $51,770 per year, that translates into over $17 billion in labor costs. If anything, these people can be “freed up to do more value-added activities” instead of “pipetting”.

It was Steve Jobs who once said that “the biggest innovations of the 21st century will be at the intersection of biology and technology”. When you see what direction laboratory automation is taking us, it’s hard to disagree with that statement.


Leave a Reply

Your email address will not be published.