Pathogen evolution, selection and immunity


Short course taught by Erick Matsen and Trevor Bedford for MITII 2017 Summer School on Modeling Immunology.

An introduction to immunity from an evolutionary perspective. We will introduce basic notions of molecular evolution, phylogenetics, and evolutionary models which will set the foundation for modeling pathogens and antibody-making B cells. We will additionally introduce multistrain compartmental models and introduce the diversity generating mechanisms of B and T cells, and survey repertoire analysis.


Course introduction

  • Placing pathogen / immune interactions into the larger ecological / evolutionary context
  • Strain dynamics and population turnover

Multistrain models and serology

  • Multistrain compartmental models
  • Exercise: Dynamics of a two-strain SIR system
  • Serological binding and neutralization data
  • Antigenic cartography
  • Exercise: Antigenic map
  • Original antigenic sin

Selection and evolution in B cells

  • General outline of adaptive immunity
  • VDJ recombination
  • Germinal center formation and cycling
  • Somatic hypermutation and mismatch repair
  • A bit of antibody structure

The coalescent and selection

  • Introduction to Kingman’s coalescent
  • Effective population size and demographic inference
  • Exercise: Skyline plots
  • Effects of selection on tree topology
  • Introduction to Wright­-Fisher model
  • Tree­-based tests of selection: Tajima’s D, etc…
  • Exercise: measles vs flu, selection and immunity

Introduction to phylogenetics

  • Sequence alignment
  • Types of phylogenetic inference methods
  • Exercise: Parsimony reconstruction
  • Intro to likelihood
  • Exercise: Likelihood surface
  • Likelihood-based phylogenetics
  • Intro to Bayesian methods
  • Exercise: Prior and posterior
  • Markov-chain Monte Carlo

We thank Sarah Cobey for generously allowing use of materials developed for her SISMID module.

All contents including slides, course materials and code are copyright 2015-2017 Sarah Cobey, Erick Matsen and Trevor Bedford. All slides / course materials (files ending in .html and .md) are licensed under Creative Commons Attribution 4.0 and all code (files ending in .py and .ipynb) is licensed under an MIT License.