A mathematician, computational biologist and former mountain guide, Dr. Frederick “Erick” Matsen is all about trees.
The tall swaying ones that line the forest paths he hikes with his data scientist wife and friends — and the dense, data-laden evolutionary, or phylogenetic, trees he infers with his research colleagues at Fred Hutchinson Cancer Research Center.
Matsen uses biological data to develop computer algorithms that help scientists better understand our biological systems and their relationships with each other.
He does this by analyzing data from the sequenced DNA and RNA of viruses, immune cells and other evolving systems. By understanding evolution, especially the evolution of pathogens and immune resistance in their hosts, we can better understand ourselves and our world. And better take on the culprits that can plague the human organism, like cancer and COVID-19.
“The sequences in many data sets are related to one another via evolution,” he said. “You need to take the structure of this evolutionary tree into account when you’re analyzing the data. If you don’t do that, then major problems happen.”
Over the 10 years he’s been at the Hutch, Matsen has used his mathematical chops to help his colleagues better understand viruses and how our “amazing” immune system responds to them.
“The immune system develops customized ways of recognizing foreign molecules,” he said. “It’s truly a miraculous process. It basically comes up with custom receptors to recognize specific other molecules. And the diversity of these things is just unreal. There are tens of millions to almost a billion unique receptors depending on the cell type. It’s incredible how many different things your body needs to recognize.”
Matsen sees this incredible diversity via high-throughput sequencing of our immune system’s top disease-fighting soldiers, its B and T cells.
This work includes a long-term collaboration with infectious disease researcher Dr. Julie Overbaugh to improve B cell response to HIV.
“Julie had antibodies from children infected with HIV that learned to recognize a broad spectrum of HIV variants in a short time,” he said. “Adults take a very long time to get good HIV antibodies that can block a broad spectrum of HIV strains. That’s an example of the sort of thing we care about.”
Mathematician on a mission
Matsen came to cancer and virus research via mathematics.
The Howard Hughes Medical Institute-Simons Foundation Faculty Scholar loved math as a kid and assumed “I’d just do math until I found something I liked better.” As it turns out, he never did. Not after trying pre-med or physics or mechanical engineering or even computer science.
“None of it kept my attention very well,” he said. “Math is a bit of an addiction. I just love it too much.”
Matsen studied math in France on a Fulbright scholarship, then in Harvard’s Math Program where he intended to become a research mathematician. But then he encountered the heady combination of evolution and math, a scientific sweet spot where he could use his math and modeling skills to help people fight disease, and he was hooked.
These days, he said he wants to be “pushed every day to think about the problems that biologists care about. Now more than ever. I want to be serving the researchers’ needs and that of the public.”
“It’s important as an institution to remember so much of our research funding is tax money,” he said. “It’s a gift from the American people to us as researchers and they are trusting us to do useful things with it.”
Going forward, Matsen and his lab mates will be working at the Hutch’s Lake Union Steam Plant, specifically designed for efficient and effective collaborations between clinical researchers, data scientists, evolutionary biologists, infectious disease experts and others.
Computational methods to ‘prep for the next outbreak’
Currently, Matsen is collaborating from home, working to develop methods to guide outbreak control for viruses such as the new coronavirus and its infection, COVID-19.
He is inspired by the work of genetic epidemiologist Dr. Trevor Bedford, who’s been tracking the path of COVID-19 and its various mutations for nearly a year via the open science collaboration, Nextstrain.
“Nextstrain is amazing and it’s been a really useful epidemiological tool for guiding interventions,” Matsen said. “But the need to be a real-time platform constrains the types of algorithms it can use. For example, it cannot use complex epidemiological models with many sources of data or integrate out uncertainty in the phylogenetic tree.”
Current implementations of these more complicated methods of analysis can take a month to run and “in the course of an epidemic a month is an eon,” Matsen said.
The Matsen group wants to deliver new algorithms for these methods that scale significantly better to large data sets.
“Our methods aren’t yet ready for COVID-19 – but we’re working on computational methods that will allow us to be prepared for the next big outbreak,” he said. “We want to make it possible for people to use these more sophisticated methods in the course of an epidemic.”
“The immune system develops customized ways of recognizing foreign molecules. It’s truly a miraculous process.”
In another collaboration with Overbaugh and Dr. Helen Chu, Matsen is helping understand the body’s antibody response to SARS-CoV-2, using a technique called phage-immunoprecipitation sequencing, a type of comprehensive analysis.
“We’re using a technique called PhiP-Seq that basically asks the question, ‘Hey, blood, what do you recognize on the SARS coronavirus? What are the things your antibodies recognize?’” he said. “‘Do you recognize this stretch of the virus? Or that stretch of the virus?’”
This work is essential now that novel viruses can go global in a matter of days.
“You can get on a plane and within 12 hours, you can take a virus from anywhere to anywhere,” Matsen said. “And that’s new in the last 50 years. If we’re talking about the longer course of history, we are going to see more global outbreaks.”
Public health goes viral
The spread of coronavirus has helped pique public interest in Matsen’s chosen field.
“It’s not like I’ve been working in an obscure field, but it is interesting to be working at a time when phylogenetic trees are mentioned in the New York Times on a regular basis and my in-laws want to talk about serology tests and antibodies and the evolution of viruses,” he said.
“My group has been working on these things for a decade at the Hutch. It’s nice that people now understand why we would care about these things.”
Matsen cares deeply about many things: working for the public good, politics, science, creativity, the outdoors, wonderment and yes, laughter and having fun. The logo for his lab is a small child holding a pencil riding a robot chicken.
He commissioned the artist to create it after he saw a similar drawing by her in a children’s book he was reading to his oldest daughter. His youngest child was born just weeks ago.
Yes, there’s yet another important tree in Matsen’s life – a newly-branching family tree. And he can’t wait to share with his daughters, still wee shoots, “all the beautiful things that happen in math.”
― By Diane Mapes, November 17, 2020