A direct result of using better technology leads to an increase in high quality data gathered.
A hallmark of life is its complexity and unpredictability. In biology laboratories worldwide, scientists adopted model organisms for experimentation. Escherichia coli is the model bacterium, Saccharomyces cerevisiae is the model eukaryotic cell and Arabidopsis thaliana is the model plant. For decades researchers have used these same strains to get repeatable results. But there is one lab organism that cannot be relied upon in its predictability: humans.
Biology is rapidly becoming an engineering-focused interdisciplinary field with synthetic biology and systems biology being recent examples of the turning tide in biotech research. Increasingly precise or high-throughput experiments need specialized hardware and software solutions. A direct result of using better technology leads to an increase in high quality data gathered. In recent years, biologists, software engineers and automation specialists have, together and separately, developed processes for designing and automating laboratory experiments, as well as analyze the mountains of data gathered with a fine-toothed comb. Welcome to the age of computer-aided biology.
Harder, better, faster, stronger
Automation has several advantages. For starters, the robot isn’t going to forget if it added a reagent already. Liquid handling is consistent unless programmed otherwise. Machines log every action performed, to the second, and repeats the action precisely for every sample. There is also the option of recording additional data like ambient temperature and humidity, which can be analyzed alongside the experimental results. This level of precision is just one of the advantages.
Another big advantage is throughput. Robots can be equipped with a single pipette tip or hundreds. Machines – and this probably won’t come as a shock to you – don’t need much sleep either. For designing new experiments, throughput is not as important. But for consistent, high-quality production of small molecules like peptides and DNA, throughput is everything. A plethora of liquid handling companies are finding growth in synthetic biology, a field closely tied to the computer-aided biology. Robotics from US-based Hamilton and Opentrons, and European Analytik Jena are favorites in the field, but many options are available depending on the application.
Robots are just half of the story. The software controlling how experiments are designed, how robots interact, and how the data are analyzed is just as important. Riffyn and Synthace have developed experimental design software that can in turn program the automation equipment. Other aspects of design are also taking advantage of artificial intelligence/machine learning. Desktop Genetics, for example, has developed a CRISPR design software based on their library of constructs. LabGenius also have an AI platform called EVA to aid in their therapeutic peptide design.
Data analysis is another important factor. While giants like Pfizer use IBM’s Watson to aid in their drug discovery, you need to turn to innovators in the field for more bespoke tasks. Eagle Genomics, a UK company recently branching into the US, have a platform for analyzing data sets on genomic and microbial scales, demonstrating interactions that would otherwise be near-impossible for a single human to identify alone.
Meeting of minds
Computer-aided biology at first glance appears like scientists are being replaced by robots just as weavers were replaced by power looms in the Industrial Revolution. The reality is completely different. There will always be a need for scientists to perfect new methods on the bench before it gets taken to automation. At Ginkgo Bioworks in Boston, software engineers work in tandem with biologists to develop automation platforms to create new opportunities in biotech.
“Automated processes are essential when it comes to collecting the large scale, high-quality data that’s needed to understand complex biological problems, and biotech has used automation for high throughput experiments for a long time. At Ginkgo, we see an opportunity to use software and automation to unlock that power of biology in a flexible and scalable way, to enable many different companies and many different industries to program cells,” says Jennifer Young, a software engineer at Ginkgo.
This meeting of fields requires a diverse skillset. Some PhD training programs recognize and support bringing disparate fields of software and biology together, but it will be some time before it becomes the norm. To find a way around this, a computer-aided biology community was launched in October.
“The Computer-Aided Biology Community was initiated after it was identified that there was a gap at the intersection of biology and technology for a focussed community of people that are looking to drive the field forward. Although we (the industry) would all benefit from something like this, to date it has not been realized. We are ready to bring the right people together and facilitate this!” says Fane Mensah, Head of Scientific Community.
The group stems from Synthace’s white paper on computer-aided biology, who are keen to get experts from broad areas at different career stages to bring energy and expertise to the table. Mensah acknowledges that widespread adoption of computer-aided biology will not come smoothly.
“However, by showcasing case studies of how technologies have affected biology and documenting these in publications the adaptation will increase. Furthermore, a breakthrough in the field in for example drug development would also give a monumental boost,” Mensah adds.
Companies like LabGenius, with an AI and automation platform, and Ginkgo are likely to be the first to give the field a “monumental boost”. Young finds the challenge ahead an exciting one, but also foresees issues in parsing biology into software.
“As a software engineer, I think the most interesting [thing] is that biology is so messy, complicated, and context-dependent. Abstracting biology to fit into databases and models that still make sense and allow for complexity is really difficult. It’s my favorite part of my job, but it can make it really difficult to build software that is easy to use yet allows for complexity that computational biologist require,” she says.
Companies are already recognizing the power that incorporating software or automation into their workflow can bring. Ginkgo has partnered with start-up accelerators Petri and YCombinator to help new companies take full advantage of automation. Synthace, besides spearheading the Computer-Aided Biology Community, has made other strategic partnerships. Mensah is convinced the people are there to build computer-aided biology into something great.
“I would encourage people to join the community. Not just to learn more about the field of computer-aided biology, but also to have their ideas, insights, and expertise shared with a focussed group of academic and industry professionals from a variety of backgrounds.”
