Research

I’m interested in the phenomenology of evolution and the immune system. What forces guide the evolution of viruses and other pathogens, and how does the immune system effectively respond to them? My research emphasizes data and measurement: how can we learn about these processes from available data, and how can we make predictions that can be tested experimentally? The ultimate goal of this work is to help improve the ways that we prevent and treat disease.

Virus evolution

Highly mutable pathogens like HIV and influenza are a major problem for human health because they can evolve resistance to immune responses that would otherwise control infection. This is why influenza vaccines need constant updates, and one of the key reasons why massive efforts have yet to produce a successful vaccine against HIV.

We want to build a quantitative, predictive understanding of pathogen dynamics at the level of individual hosts and whole populations.

Current projects:

Inferring selection acting on HIV and influenza from time-series genetic sequence data and using this information to predict function and future evolution
Understanding the genetic composition and dynamics of the “latent reservoir” of quiescent HIV-infected cells, which present a major barrier to HIV cure efforts

Key papers:

Relative rate and location of intra-host HIV evolution to evade cellular immunity are predictable
Barton JP, Goonetilleke N, Butler TC, Walker BD, McMichael AJ, Chakraborty AK
Nature Communications, 2016 [journal link] [pdf] [si]

Paired quantitative and qualitative assessment of the replication-competent HIV-1 reservoir and comparison with integrated proviral DNA
Lorenzi JC, Cohen YZ, Cohn LB, Kreider EF, Barton JP, Learn GH, Oliveira T, Lavine CL, Horwitz JA, Settler A
Proceedings of the National Academy of Sciences, 2016 [journal link] [pdf] [si]

Scaling laws describe memories of host-pathogen riposte in the HIV population
Barton JP, Kardar M, Chakraborty AK
Proceedings of the National Academy of Sciences, 2015 [journal link] [pdf] [si]

Innate immunity

The innate immune system is our first line of defense against invading pathogens, and it also helps to fight cancer and coordinate adaptive immunity. Some innate immune cells use an extremely diverse array of germline-encoded receptors to discriminate between healthy cells (“self”) and those which are foreign or infected (“nonself”) as well as the direct detection of microbial products. How do innate immune cells, whose receptors are fixed and immutable, manage to protect us from a broad range of continually-evolving pathogens? And more broadly, what fundamental principles underlie this mode of pathogen recognition, which is so different from the paradigm of adaptive immunity?

Statistical inference

We live in an exciting time in the development of experimental methods to probe genetic evolution. Consider this dramatic example documenting viral spread during the recent Ebola outbreak in West Africa. It is also possible to record evolving populations of viruses within chronically-infected hosts, analyzed in our own work here.

Recently-developed techniques have also enabled us to look at cellular heterogeneity in fantastic detail. This heterogeneity clearly plays a key role in some aspects of the immune system, but little is understood quantitatively.

One of our goals is to develop methods to efficiently learn from this kind of data, and to use it to predict new behavior.

Current projects:

Inference of selection and other evolutionary parameters from time-series genetic sequence data
Improved maximum entropy inference methods for correlation data using methods from machine learning

Key papers:

ACE: adaptive cluster expansion for maximum entropy graphical model inference
Barton JP^c, De Leonardis E, Coucke A, Cocco S^c
Bioinformatics, 2016 [journal link] [pdf] [si] [code]

Ising models for neural activity inferred via selective cluster expansion: structural and coding properties
Barton J^c, Cocco S^c
Journal of Statistical Mechanics: Theory and Experiment, 2013 [journal link] [pdf]