How Do Viruses Evolve?

Pew biomedical scholar explains how new ones emerge—and vaccines are developed

How Do Viruses Evolve?

As coronavirus cases continue to rise across the globe, scientists and policymakers are fervently working to mitigate this public health threat. The first cases involving a new coronavirus— SARS-CoV-2, or the disease named COVID-19—appeared in Wuhan, China, in December 2019, but the virus has spread rapidly since, raising significant concerns about its implications for human health.  

The latest known to infect humans, SARS-CoV-2 is in the same family as the coronaviruses that caused severe acute respiratory syndrome (SARS) in 2003 and Middle East respiratory syndrome (MERS) in 2012. Although the current fatality rate is lower than for MERS or SARS, this disease has already spread to more people and caused more deaths. At the same time, other viruses such as influenza—known as the flu—continue to evolve, adapt, and infect millions of people each year, making the ability to develop effective vaccines critically important. 

Courtesy of Marta Łuksza

In her work, Marta Łuksza, a 2019 Pew biomedical scholar and computer scientist, explores how immune interactions drive the evolution of the flu virus. Łuksza is also an assistant professor at the Icahn School of Medicine at the Mount Sinai Health System in New York. She talked with Pew recently to help put viral spread into context. This interview has been edited for clarity and length.

Q. What is a coronavirus?

A. Coronaviruses are a family of viruses that cause symptoms of respiratory illness in humans, similar to the flu or the common cold. Often found in bats and other mammals or birds, coronavirus can prove dangerous when transmitted between animals and people.

Q. How does a virus first emerge?

A. A virus may first emerge in humans after a chance interaction with an animal host, during which a person becomes infected. In many cases, viruses only transmit from animals to humans but cannot be transmitted from one human to another. In rare cases, however, the virus can survive transmission among people.

Doctors and other health professionals first identify a new virus after testing for known illnesses and finding no match. If they cannot assign the virus through testing to known protein markers or genetic material from a particular virus and there is a growing number of similar cases, this may point to something new, such as the recently identified SARS-CoV-2 virus.

Q. Why do some viruses seem to spread more widely than others?

A. Both biological and demographic factors can facilitate viral spread. Biologically, if a given virus is capable of infecting the body through accessible entry points, such as epithelial cells in the nose, it can enter the respiratory tract and spread relatively easily. When a person coughs or sneezes, the virus can circulate in the air and on surfaces. Viral spread can also depend on how fast a virus is able to replicate itself, thereby dispersing to other parts of the body or to new hosts. Demographically, dense populations where people live in close proximity are more likely to experience rapid viral spread than populations that are sparse.

Q. Why do people respond differently to viral infection?

A. People’s immune systems have memories of prior infections that affect how they will respond to a virus. To prevent infection once the body has been exposed to a virus, it produces proteins called antibodies that identify and neutralize potential threats. For instance, when people contract the flu, they have likely built up some degree of immunity due to previous exposure, including the flu vaccine. In the case of a new coronavirus, the immune system has not seen this virus before, and the adaptive response is slower.

Q. What is the process for developing a new vaccine?

A. First, researchers need to isolate the virus and identify its antigens—viral proteins that serve as the best vaccine targets—against which the immune system is likely to create antibodies to defend the body from infection. However, each virus is different and requires further growth and testing in a lab, where a range of production technologies are used. There are currently three platforms for development of the flu vaccine, including virus growth in chicken eggs, in mammalian cells, or created synthetically from the DNA sequence of the vaccine candidate strain. For a vaccine against a newly emerged pathogen, the next step is clinical trials, during which the safety, including possible side effects, and effectiveness are tested.

For many pathogens, for example the measles virus, the vaccine does not have to be modified in the future and remains effective over time. But other pathogens, such as the flu virus, have the ability to escape the vaccine’s recognition by acquiring new mutations in its antigens. This is why a biannual evaluation of the flu vaccine is critical—to update the antigens contained in the vaccine. 

Researchers, including myself, work with organizations like the World Health Organization and participate in biannual consultations to help select flu vaccines for the Northern and Southern hemispheres. Based on publicly available genetic sequence data on the flu from individuals   around the world, researchers re-create the evolution of viruses to see which mutations have occurred and how frequently they are circulating. Antigenic data—information related to antibodies that are triggered in the body by exposure to different protein markers on a virus—also help labs to characterize viral proliferation and how well vaccines block the virus. This is critical to help public health officials determine how infectious disease is dispersing globally and across continents.

Q. Can you tell us more about your research?

A. I focus primarily on the evolution of the flu virus and cancer, which are both affected by interactions with the immune system. In my role as a computer scientist, I work to develop models and software tools that aggregate data to determine a virus’s fitness advantage—the conditions under which a virus escapes an immune response.

For the flu vaccine, this allows us to better predict which of the co-circulating strains and mutations will be most common from one season to the next. My team has pioneered a class of forecasting models to better describe the mechanisms of immune recognition. These models also evaluate how the history of different viral strains at a given period of time shapes the future landscape for a virus’s antigenic escape, or when the immune system can’t recognize nor eliminate an infectious agent. We currently focus on the interactions within the flu hemagglutinin and neuraminidase proteins, which are responsible, respectively, for the initiation of an infection and then viral replication and spread.

Another exciting project I’m involved in is developing a universal flu vaccine, which would target parts of a virus that do not change or mutate over time. Once we successfully identify target regions for this vaccine, it may be able to cover many more strains of the flu virus and save thousands more lives each year.

The front facade of the Supreme Court of the United States in Washington, DC.

Agenda for America

Resources for federal, state, and local decision-makers

Quick View

Data-driven policymaking is not just a tool for finding new solutions for emerging challenges, it makes government more effective and better able to serve the public interest.


States of Innovation

Data-driven state policy innovations across America

Quick View

Data-driven policymaking is not just a tool for finding new solutions for difficult challenges. When states serve their traditional role as laboratories of innovation, they increase the American people’s confidence that the government they choose—no matter the size—can be effective, responsive, and in the public interest.

Latest From Our Biomedical Programs