Expanding the Potential of Directed Evolution Using Python-Adapted Liquid Handlers
June 8, 2022
Emma Chory, Ph.D., gives directed evolution a robust, data-responsive boost with her high-throughput, automated platform, Pyhamilton. Her work empowers existing hardware for new applications that range from biomanufacturing to fundamental biology.
Pyhamilton enables researchers to harness automation for something other than replacing a scientist, performing something faster, or simply streamlining a process,” says Chory, a postdoctoral fellow in the Sculpting Evolution Group at Massachusetts Institute of Technology (MIT; Cambridge, MA, USA). “With Pyhamilton, scientists can answer questions that would not have been feasible no matter how many hands you had to do an experiment. It can be used for applications that we haven’t even imagined – experiments that people shied away from because throughput was a limiting factor in implementation.”
Chory and her co-authors’ innovative Pyhamilton platform is an open-source integration of the Python software language and robotics from Hamilton Corp. that expands the experimental repertoire of liquid-handling systems. The platform can double the speed of automated bacterial assays over current software, execute complex pipetting patterns to simulate population dynamics, and incorporate feedback-control to maintain hundreds of remotely monitored bacterial cultures in log-phase growth without user intervention.
Most recently, Chory and her colleagues have leveraged Pyhamilton to make high-throughput combinatorial drug synergy screening more accessible to scientists and clinicians. They hope that enabling scientists to more test new drug combinations could lower the dosing and toxicity of cancer therapies.
To showcase Pyhamilton’s capabilities, Chory and her co-authors, Erika DeBenedictis, Ph.D., Dana Gretton, Brian Wang, Ph.D., Stefan Golas and principal investigator Kevin Esvelt, Ph.D., recently published phage-and-robotics assisted near-continuous evolution (PRANCE), which enables scientists to answer fundamental questions in evolutionary biology with automation. Though many evolutionary biology phenomena have been investigated in whole-organism evolution, it has been a challenge to apply these principles to the directed evolution of gene-encoded biomolecules within the laboratory. These capabilities have implications for the understanding of pathogenic escape mutations relevant to global health. Chory’s team demonstrated the ability to systematically explore evolutionary outcomes across many populations.
The work represents a new paradigm for the autonomous engineering of biomolecules and enables previously inaccessible, well-controlled scientific experiments to highlight the fundamental principles of gene-level evolution.
“Directed evolution historically has been a bit of an artform,” Chory explains, describing how typical phage-directed evolution experiments are so labor-intensive and know-how heavy that previously published papers on the topic frequently lack controls. “Experiments are run one at a time by a solitary graduate student or postdoc who knows where things are likely to go wrong, pauses the process at that precise time and makes necessary adjustments,” she continues. “Pyhamilton offers us an opportunity to turn continuous directed evolution into an engineering discipline.”
Learn more as Emma Chory unpacks Pyhamilton’s innovative toolkit, which paves the way for more experiments, not fewer lab members.
Multi-Tasking Inside and Outside the Lab
From early childhood, Chory knew that engineering was in her future as she watched her mechanical engineer dad work on helicopters. Curious and energetic, she always tackled tasks by two – working and going to school, listening to lectures and crocheting, exploring science instead of sleep.
“As an undergraduate, I brought knitting to lectures because I would find myself less likely to daydream about projects or homework if I was busy knitting,” says Chory, who uses crocheting now as an effective brain organizer. During the COVID-19 pandemic lockdowns in 2020, Chory’s needles reduced stress and provided life-balance as she generated everything from a life-sized microscope for a friend's daughter (she wants to be a scientist), to a baby-sized Cadillac for her cousin's newborn (the baby's parents are in the automotive industry) and even a robot for another friend’s son.
“Every now and then I get quirky requests for things that I have to engineer in my head to get the proper geometry of the shapes,” she says. “I had many requests for coffee-cup-shaped dog toys, so I started an Etsy store to sell them.”
Accomplishing several tasks at a time is simply her modus operandi. Chory worked her way through an experiential learning program while earning her undergraduate degree in chemical engineering from Northeastern University(Boston, MA, USA), earning 24 months of full-time work experience before graduation. One of her memorable assignments during this time was with Jay Bradner, M.D., at theDana-Farber Cancer Institute(Boston, MA, USA), which piqued her interest in life sciences research.
“I found myself wanting to go into the lab in the middle of the night to run an experiment!” she says. “Jay is an incredible mentor and phenomenal science advocate who has been a great support to me. My work in epigenetics at Dana-Farber developed an interest in how the complex regulation of our genes function. Because of this work, I decided to go to graduate school to study just that.” She traveled West to earn her master’s and doctoral degrees in chemical engineering from Stanford University (Stanford, CA, USA), where she pursued studies in epigenetics and chromatin biology in the laboratory of Gerald Crabtree, Ph.D.
“At that time, I didn’t feel that there were many engineers working in this space,” Chory continues, describing her scientific career as following an unconventional trajectory. “Epigenetics is a complicated and intimidating field that can create barriers for more traditional engineers to move into.”
At the end of graduate school, a paper on phage-based evolution published by her current advisor Kevin Esvelt, Ph.D., caught her attention because of her interest in drug discovery. Fascinated by the research, she contacted Esvelt to learn more.
“Kevin told me they wanted to use robotics to speed up the process of using bacteria to produce interesting molecules – that really hit a nerve with me. As a chemical engineer, I knew we had this huge world of synthetic biology to take advantage of, but some of the limitations for using bacteria to engineer proteins still comes down to only having so much of any given molecule that you want to use. Without miniaturization and a streamlined process, some methods are simply impossible” Chory says. She explains that in industry labs, sophisticated tracking techniques make it easier to produce molecules and tweak them, making real-time changes and quality checking experiments throughout the process. “But there wasn’t a way to do that level of tracking with this powerful, albeit cumbersome evolution system,” she adds.
Because of her background in high-throughput screening (HTS) from the Bradner and Crabtree labs, she knew that the speed and data accumulation of HTS offered Chory’s team an edge in developing PRANCE using Pyhamilton. “We could use the numbers game. The sheer amount of experiments we conducted using HTS and robotics to learn about evolution gave us more statistics and accuracy than previously have been available,” she says.
“As an academic group, we had the freedom to devote time to develop these flexible capabilities,” Chory continues. “Other groups, biotech startups, for example, are motivated to work with robotics in a way that is more user friendly and open sourced, but I think it’s tricky for those companies to devote manpower to building a system when it’s not a necessary objective – at least not compared to getting a drug to clinic or meeting a benchmark you need to meet for shareholders.”
And that brings us back to Chory’s current work at MIT. She believes that her team’s platforms are beneficial to many other life sciences organizations. “Python is elegant enough to speak to computer scientists, but accessible enough that biologists can learn to work with it,” she explains, noting the number of SLAS2022 exhibitors she noticed that were showcasing their Python-adapted products, which with Pyhamilton they could now easily integrate with Hamilton robots. Chory adds that exploring the exhibition was a thrill for her and lab technician Stefan Golas.
“We were like kids in a candy store!” she says. “We’re always on the lookout for out-of-the-box components we don’t have to build ourselves. A few parts of our system were tinkered using things we found on Amazon, but that makes it difficult for others to reproduce. If we can incorporate ready-made parts, it makes version 2.0 that much easier. This is the heart of our Pyhamilton platform: We don’t replace anything that Hamilton does, we simply make it easier to interface with the robot to enhance what their machine already offers.”
Chory would like to see their open-source Pyhamilton platform adopted into the infrastructure that already exists at many universities. “High-throughput drug screening centers often use these same robots. It would be phenomenal to see core facilities adopt Pyhamilton to tackle more than compound screening, PCRs, or high-throughput cellular assays. These are sophisticated robots – they can do more than pick up a plate and put it down!” she says.
Chory acknowledges that while people are excited about the platforms, “change is slow – especially in academia. It’s an odd thing, but I get a sense that academic labs have a mindset that automation replaces graduate student work. One robot does not equal 10 grad students – but now, robots can empower researchers to tackle big questions in ways they never dreamed possible!”
She hopes to help facilitate that change with her next move. Chory will be starting her lab at Duke University as an assistant professor of biomedical engineering in the Center for Advanced Genomic Technologies (CAGT) in the summer of 2023. The Chory lab will be developing new therapies and tools to study gene regulation and using robotic directed evolution