University of Southern California: Sophisticated math and computing — and a healthy dose of collaboration — tackle questions about genetics

Inscribed within our cells could be our future, a story written before we are even born.

Within each cell of the human body, a person’s genes contain instructions that influence traits like height, eye color, and even whether or not we like the taste of cilantro. And variations in our genes can alter our risk for certain diseases, such as breast cancer and schizophrenia.

Of course, it’s not quite this simple; many factors play a role. Some genes are only “activated” under certain environmental conditions. Upbringing or cultural forces can overcome genetic predispositions, in many cases, though scientists aren’t certain to what extent.

Researchers at the new Department of Quantitative and Computational Biology (QCB) at the USC Dornsife College of Letters, Arts and Sciences are using advanced math and computer science to answer some of the important questions surrounding our genes, and the destiny they foretell.

The culture in QCB emphasizes advanced computing and collaboration, aiming to foster an ideal environment for discovery. At other universities, many of these scholars would be the only computational researcher in their department. Here, they’re surrounded by them.

“In QCB, if I run into a computational problem, I can just walk down the hall and talk to someone,” says Michael “Doc” Edge, assistant professor of Quantitative and Computational Biology. “I can actually grow my work this way.”

Finding the missing link

Michael “Doc” Edge. (Photo: Courtesy of Michael Edge.)

In 2003, scientists sequenced the bulk of the human genome for the first time, revealing all the letters of code in our DNA. It cost about a billion dollars and over a decade to complete. Since then, advances in technology have cut down the effort considerably. It would take mere weeks and around $1,000 to do this same work today.

The technology that can currently produce a genome at this price does have drawbacks: Some 3% of the human genome cannot be interpreted using this technology. Yet, these missing sections may contain genetic variations that could tell us much about ourselves.

“If we want to predict how someone’s genome will affect how they respond to a drug, or how likely they are to develop a particular disease, if we’re missing some fraction of it because we’re using this shorter-read technology, then we’re not seeing the full picture,” says Mark Chaisson, assistant professor of quantitative and computational biology.

Chaisson’s lab is developing powerful, intricate methods to add that missing percentage back in. Once the gap is filled, Chaisson and his students can find more information about genetic variation and how it might contribute to inherited diseases.

Much of his work is particularly focused on structural variants, which occur across a large swath of the DNA sequence rather than just a single, small point.

Much of his work is using large-scale studies and massive databases, which means the powerful computers on-hand at QCB are vital to his research — as are his colleagues.

“Everything that we produce is either the result of a person sitting at the computer themselves, or chatting with another person here and learning some new skill or sharing ideas,” says Chaisson.

Complex genetics

Over in Edge’s lab, he and his students are researching “complex traits,” traits that arise from a combination of both genetic variation and behavioral factors and environmental conditions.

For instance, a person’s height is influenced by hundreds of genetic variants, and it’s how those variants blend together that dictate how tall someone grows. Height is also heavily influenced by nutrition, however. A person with genetics that might normally have them pushing 6 feet could end up shorter due to inadequate nutrition.

Edge also studies issues surrounding genetics and privacy. A recent rise in home consumer genetics tests, used to explore one’s ancestry or propensity for certain diseases, has resulted in vast databases of genetic information

Law enforcement officers and consultants are also increasingly searching a subset of these databases using DNA taken from crime scenes, a method dubbed “forensic genetic genealogy.” California’s Golden State Killer, who murdered at least 13 people and had been hunted since the 1970s, was finally identified in 2018 thanks to this sort of search.

Edge has written both about the power of this kind of detective work and about the privacy concerns it raises. A person who uploads their personal genetic test to an ancestry database unwittingly uploads information about all of their relatives who share their genes.

“When you upload a test, it doesn’t just affect you; it potentially affects a bunch of people you’re related to, many of whom you’ve never even met,” says Edge. The rules around these searches are currently enforced only by the individual companies. Just two states have laws in place about forensic genealogy.

“The majority of the people I’ve talked with don’t want to outlaw searches. They’d just want to think through all of the implications before allowing unfettered access.”

Finding family

Despite the field of genetics’ incredible potential to inform us about disease and human history, it also has a somewhat uncomfortable origin story.

Since its inception at the turn of the 20th century, genetics has often been intertwined with eugenics, the effort to breed perceived flaws out of the human race. Genetics research was often pursued or co-opted by eugenicists eager to eliminate “undesirable,” inherited traits through sterilization or anti-miscegenation laws.

“The controversy in genetics stems from racism that was brought into our field quite early on. Early motivation for doing genetic studies was to show that there are differences between assumed races, thus making some groups superior to others,” says Jazlyn Mooney, Gabilan Assistant Professor of Quantitative and Computational Biology. “Now our goal is to push our work into the right direction, while acknowledging this complex history still exists.”

Mooney is a new arrival at USC Dornsife, having set up her lab in the spring of 2022. She focuses on population genetics — the genetics shared within a particular group of humans, how traits are passed on and expressed, and how diseases might be inherited.

She’s one of many new geneticists who are intent on moving the field out from under the shadow of its past. Data sets used in genetics research and stored in ancestry databases are primarily European, with little representation of non-Europeans such as African, African American or Indigenous groups.

Mooney is particularly interested in understanding the history and diversity of non-European and admixed populations. Admixed populations, like African Americans, occur when people possess genetic ancestry from two or more distinct sources, such as Europe and Africa.

This work is personal for Mooney, whose father is African American and mother is Hispanic and Native American. She can trace her mother’s side of the family back to Spain all the way to the 1500s, but has little information about her father.

“We want to do good things for human health, but if you don’t have non-European people in your data then those people are at a disadvantage,” says Mooney.

“That’s the big thing when we’re thinking about the future of human genetics. How do we incorporate diverse groups into our data sets and also make sure that their community sees a tangible result from the science?”

For a scholar like Mooney, who is just embarking on her career, the QCB department is an excellent place to grow. “We have a lot of young faculty members, which is different from many other genetics departments,” says Mooney. “Remo Rohs, as the department head, really has a vision for the younger faculty members. It’s a great place to be.”