Earlier this month, major tech consulting company Cambridge Consultants announced a partnership with Synthace, a small, U.K.-based software startup. According to the press release, the collaboration is designed to “unlock the full potential of biology” by increasing adoption of Synthace’s operating system and programming language for biological experiments.
The idea of automating wet lab work—such as pipetting microliters of liquid, pouring gels, or mixing reagents—is not new to biology. Yet adoption of expensive robotic systems has been largely limited to pharma companies that have pockets deep enough to afford them. Thus, the manual, painstaking work of most wet labs continues.
In sharp contrast to the complexity of even a simple biological cell, “biotechnologists engage with a handheld pipette, an Excel spreadsheet, and a pen and paper,” says Tim Fell, CEO of Synthace. And because of that tradition, experimental methods are recorded in text files, and results are stored in data sets whose formats vary from lab to lab, making them hard to combine.
Several young companies are now working to bring biology-specific software and programming languages to the lab in order to automate such tasks. They hope not only to speed up experiments but also to improve their reproducibility and standardize data gathering so that data can be analyzed using machine-learning techniques. “Biology is so complex, our brains just can’t see all the interactions, so we’re going to need machine learning,” says Fell.
“The future is really about generating larger data sets and having access to more powerful computational tools,” adds Max Hodak, founder and chief technical officer of robotic cloud lab startup Transcriptic. “I’m not sure the way most academic research labs [currently operate] can lead to that improvement.”
Credit: Synthace
Synthace, a 28-person company established in 2011, created a platform called Antha, an operating system and programming language designed to allow scientists to engage with biology in a completely new way, says Fell. The software is device agnostic, so researchers can use their own favorite mass spectrometers and centrifuges in combination with the software. Using Antha to code an experiment, one can look at dozens to hundreds of variables at a time (instead of one), and there is no need to manually track all the experimental details. Instead, the software automatically records everything, including reagents, temperature, and timing, and saves the data and methods in a standardized way.
Transcriptic, a 40-person startup founded in California in 2012, provides a robotic cloud lab service that allows researchers to log into the company’s automated labs on demand. Each robotic work cell in the facility (soon to be sold commercially to large companies) is powered by custom, open-source software. Users interact with the platform via a Web interface linked to custom databases. “It fulfills the traditional role of lab [software] and the lab notebook,” says Hodak, who was recently announced as a cofounder of Elon Musk’s new venture, Neuralink.
Behind Transcriptic’s Web interface is a custom software called Transcriptic Common Lab Environment, or TCLE, which operates the infrastructure. This set of APIs interacts with the lab, checking inventory, executing protocols, even auto-formatting data. TCLE is powered by Autoprotocol, a data standard developed by the company and subsequently released as open source.
The robotic lab itself is made up of approximately 20 devices including pipetting systems and plate readers that can be combined to perform a huge range of experiments, says Hodak. “The platform is extremely open-ended.”
And here’s why these types of startups matter: Just as most of us don’t care how a printer works, a scientist using these platforms doesn’t have to track (or care about) every detail of an experiment. The software does it for her, and saves it in a way that can be stored and shared with others.
“Say I want to assemble 40 genes into an organism. Then all of the complex details [involved], be it liquid handling or using the equipment, just disappear from view,” says Fell.
Synthace offers both an open-source version of its language, and a commercial version including the OS and support. Most of Synthace’s and Transcriptic’s clients are currently Big Pharma companies, but both claim they are working their way down to smaller users. Fell hopes that as the costs of robotics systems for biology decrease, more academic centers will use the platform.
