UW BIOFAB: A force for reproducible science

The UW’s Biofabrication Center partners with Agilent Technologies in pursuit of automated, reproducible research

Undergraduate technicians perform common molecular biology tasks to enable the rapid design, construction and testing of genetically reprogrammed organisms for biotechnology applications and research. Dennis Wise / University of Washington

Key to advancing any new scientific discovery is the ability for researchers to independently repeat the experiments that led to it. In science today, particularly biology, the lack of reproducibility between experiments is a major problem that slows scientific progress, wastes resources and time, and erodes the public’s trust in scientific research.

At the University of Washington, researchers have access to the UW Biofabrication Center, or BIOFAB, a unique facility located in the Nanoengineering and Sciences building in which scientific protocols are encoded as computer programs, allowing undergraduate lab technicians to execute experiments according to detailed instructions.

“The BIOFAB is unlike any other lab on campus,” says BIOFAB founder Eric Klavins, Professor and Chair of the UW Electrical and Computer Engineering Department. “In effect, we’ve been able to automate common protocols by using software to assist our student technicians. This ‘human-in-the-loop’ system goes a long way towards improving the replicability of biological research.”

In an effort to expand the lab’s automation capabilities, the BIOFAB has partnered with Agilent Technologies Inc., a life sciences development and manufacturing company based in California’s Silicon Valley. Using state-of-the-art research equipment from Agilent, the BIOFAB will develop high-throughput workflows for common tasks of interest to members of the synthetic biology community.

Programming the biology lab

Computer programmers write code to tell a computer what to do and how to do it. For a given program, the same inputs consistently result in the same outputs.

In contrast, two biology researchers can seemingly carry out the same experiment, but get different results. This is in part because instructions for how the experiment was conducted – whether documented in a lab notebook or published in a journal – are often vague or incomplete, leaving out details that the author may not have realized impacted the experimental outcome.

Aquarium is a web-based software application that allows scientists to build executable protocols, design experimental workflows based on those protocols, manage the execution of protocols in the lab and automatically record the resulting data. Dennis Wise / University of Washington

As a computer scientist turned synthetic biologist, Klavins realized what biologists needed was a more formal way – a programming language – to define how to conduct an experiment. This led to the development of Aquarium, a web-based software application that allows scientists to build executable protocols, design experimental workflows based on those protocols, manage the execution of protocols in the lab and automatically record the resulting data.

“Aquarium provides the means to specify, as precisely as possible, how to obtain a result,” said Klavins.

When it comes to engineering biology – reprogramming cells to produce chemicals or drugs, or perform complex functions like sensing toxic compounds in the environment – reproducibility is paramount. The BIOFAB uses Aquarium to standardize various scientific workflows, generating reliable and highly reproducible results. The BIOFAB is one of a growing number of labs known as biofoundries which are committed to efficiently engineering biological systems and workflows.

BIOFAB operations are overseen by two lab managers, with a dozen or so undergraduate students executing jobs for BIOFAB clients. BIOFAB technicians perform common molecular biology tasks like DNA assembly and purification as a fee-for-service to the scientific community. Since its founding in 2014, the BIOFAB has run over 30,000 jobs for 300+ different clients at the UW and beyond.

“The BIOFAB has been absolutely instrumental in establishing and executing robust Aquarium driven protocols for a major portion of our de novo design minibinder pipeline,” said Lance Stewart, Chief Strategy and Operations Officer at the UW’s Institute for Protein Design (IPD). IPD researchers use computers to design millions of minibinders – small, stable proteins that bind with high affinity to targets of interest – that must be produced and tested in the lab. IPD uses the BIOFAB to screen minibinder candidates for protein stability and protein:protein interactions, which involves constructing yeast libraries from chip synthesized oligonucleotide genes encoding minibinder designs and carrying out large scale fluorescence activated cell sorting and next generation DNA sequencing.

“By handing off time-consuming wet lab work to our technicians, BIOFAB clients like IPD can focus more on the design and data analysis aspects of their experiments,” said Klavins.

Learning by doing

On any given day, the BIOFAB is buzzing with undergraduate technicians working together in harmony to complete an assortment of experiments for BIOFAB clients. Most technicians start working in the BIOFAB as freshman or sophomores, and for many, it’s their first real lab experience.

BIOFAB technician Nicole Roullier. Dennis Wise / University of Washington

Upon joining the lab, BIOFAB lab managers teach students basic lab skills, such as pipetting and sterile technique, and orient them to the lab. Armed with this foundational knowledge, BIOFAB technicians can begin executing a variety of different protocols by following the step-by-step instructions provided through Aquarium. Students become adept at performing complicated experimental workflows involving complex equipment through the process of doing them over and over again.

“Aquarium allows us to effectively train many students simultaneously and get them working in the lab relatively quickly,” said Aza Allen, a lab manager at the BIOFAB. “Aquarium’s technician interface makes it easy to get undergraduate students, who do not necessarily know much about molecular biology when they start, to perform experiments reliably.”

“I have learned so much beyond what could possibly be taught in a classroom setting,” said BIOFAB technician Nicole Roullier, a UW biochemistry senior. “Most undergraduates don’t have the opportunity to work with such sophisticated equipment and master advanced techniques like qPCR and next-generation sequencing (NGS). This hands-on training has built up my confidence in the lab in preparation for graduate school.”

A promising partnership

The BIOFAB provides critical automation and analytics infrastructure dedicated to enabling the rapid design, construction and testing of genetically reprogrammed organisms for biotechnology applications and research. Through its partnership with Agilent, the BIOFAB aims to offer new high-throughput capabilities that will further speed up and scale up synthetic biology research.

“We’re thrilled to be partnering with Agilent,” said Klavins. “Their support will not only accelerate the development of innovative technologies, but will help us educate students using cutting-edge equipment, bolstering our ability to prepare students for success in their own future research and career.”

“We think this is the start of an exciting collaboration,” said Kevin Meldrum, General Manager and Vice President of Genomics at Agilent. “We are pleased to be able to support researchers at the UW and the educational mission of the university through the BIOFAB. We see this as an investment in the future of our field.”

Agilent’s state-of-the-art liquid-handling robot, the Bravo Automated Liquid Handling platform will help speed up and scale up synthetic biology research. Dennis Wise / University of Washington

As a result of this partnership, the BIOFAB has acquired several valuable pieces of equipment, including Agilent’s state-of-the-art liquid-handling robot, the Bravo Automated Liquid Handling platform. While the Bravo can be used to automate sample preparation for a variety of different applications, the BIOFAB plans to initially use it to expedite its workflow for NGS. In addition to the Bravo, the BIOFAB has also acquired the AriaMx Real-Time PCR System, and the 5200 Fragment Analyzer System, a parallel capillary electrophoresis system.  

“Library preparation for high-throughput NGS is a tedious, labor-intensive process,” said Klavins. “Agilent’s Bravo will help make this workflow more efficient and reduce pipetting errors that make results less consistent, while also freeing up time for our technicians to work on less repetitive tasks. We know that there are certainly other workflows that would benefit from the use of Bravo, and we plan to engage BIOFAB users to identify which ones to pursue. We are thrilled to be able to bring this resource to the UW community, and are excited to see the compelling science that comes out as a result.”

How to build a network of pharmaceutical biofactories

By Lindsey Doermann | Originally published on the UW Chemical Engineering website

Advances in synthetic biology and biomaterials open up exciting prospects for distributed manufacturing of drugs, food products, and other commodities

illustration of pills filled with network pattern

During the world wars, American factories that ordinarily made cars and appliances changed up their production lines to manufacture guns, tanks, and aircraft in support of the war effort.

Fast forward to today when we’re facing a different fight, the adversary being a novel virus. When a disease spreads, certain drugs and their precursors can suddenly become as valuable as tanks and body armor. Is there a way re-tool pharmaceutical manufacturing, as industries did during the war, to help meet demand for valuable molecules? 

An interdisciplinary research team that includes UW engineers thinks so — by employing a synthetic-biology take on wartime manufacturing. Their goal is to design flexible, plug-and-play modules that produce valuable products within existing fermentation equipment. Put another way, breweries, distilleries, and other fermentation facilities could adapt their “peacetime” operations to start churning out pharmaceuticals in response to public health emergencies.

The group is led by soft-materials specialists Lilo Pozzo of ChemE and Alshakim Nelson of Chemistry; synthetic-biology experts James Carothers of ChemE and Hal Alper of University of Texas, and small-scale automation expert Nadya Peek of HCDE. The team recently received NSF funding to develop the technology and prototype modules for distributed chemical manufacturing. Here’s how it would work.

Many of our drugs or drug precursors derive from plants. Aspirin originally came from tree bark, opiates have been farmed from poppy flowers, and the anti-cancer agent Taxol was derived from the bark of the Pacific Yew tree. The anti-malarial artemisinin is still largely sourced from the leaves of sweet wormwood trees farmed in Asia and Africa.

But when it comes to pharmaceutical production, Pozzo points out, “plants are inefficient because the component you want to isolate may exist in very, very small quantities.” Thanks to synthetic biology (SynBio), researchers can hack this process by genetically modifying bacteria and yeast cells to overproduce desired chemicals. Think of it as a sophisticated form of brewing beer.

3d printed hydrogel lattice
A 3D-printed hydrogel lattice. Photo courtesy of Trevor Johnston/Nelson Lab.

In a February Nature Communications paper, Nelson and Alper reported a major step toward making SynBio systems portable and adaptable. They discovered how to embed genetically-modified microbial “factories” within a 3D printed hydrogel. In this configuration, they can constrain chemical production to within the gel (as opposed to dispersed in a large fermenter), run reactions continuously, and even grow more than one microbe within a bioreactor.

Now, chemical engineers can help move the ball further. In the new project, the ChemE’s will focus on developing new materials that optimize the transport of both the nutrients into the system and the chemical product out. Extracting the end product efficiently has long plagued this technology. But now, with the hydrogel framework and chemical engineers at the ready to solve some transport problems, it has new potential.

In parallel with the ChemE work, HCDE’s Nadya Peek will take the lead on designing the human-centered automation workflows for the modules. “You have to think about the human aspect of this,” says Pozzo. “Users may not have the same levels of expertise.” To be successful, the team needs to create a system that’s plug-and-play, works within existing infrastructure, and doesn’t require PhD-level training to operate.

The multifaceted project offers exciting opportunities for students, as well. Pozzo is developing design projects for ChemE undergraduates that look at the economics of different products that could be made by the SynBio systems. Students could consider not only chemical production aspects — e.g. how hard is it to make a given molecule? what are the yields using current methods? — but also the demand and market forces. The researchers view the modules as a flexible platform, and Pozzo thinks the students can help reveal just how flexible it can be.

In fact, she thinks the pharmaceutical materials are only the beginning. If you can make plant-derived drugs this way, why not other plant or even animal-based products? In the burgeoning field of cellular agriculture, microbial “factories” can replace some traditional farming operations to produce our food. In farming, animals serve as rather inefficient bioreactors to produce meat, milk, and eggs. But cells in SynBio systems are capable of producing the same things in a climate-friendlier manner, bypassing animals altogether.

Factories are traditionally designed to make one product. Reconfiguring an auto plant to produce ventilators, however critical, is a big undertaking. But think: if your factory is a cell, then reconfigurations occur on the micro scale. “You wouldn’t need to have large-scale manufacturing if you could design a cartridge, for example, that you could plug into existing fermentation equipment,” says Pozzo.

Breweries, distilleries, and commodity chemical producers exist all over the country. What if they became a network of “farms” for food or pharmaceutical products that can adapt to emergency shortages? What if microbial factories could supply some of the food we eat every day? It turns out that a big idea for a new way of manufacturing may actually be microscopic.

This work is funded through the National Science Foundation’s Emerging Frontiers in Research and Innovation (EFRI) program, which invests in high-risk, high-reward multidisciplinary research.