Code to infer antigenic maps from titer panel data.
This repo contains code to infer antigenic maps from titers panel data, building on the methods of Smith et al. 2004, and Bedford et al. 2014.
The overall goal of this project was to simulate multi-epitope antigens and explore how maps fitted to synthetic titers estimated for these hypothetical multi-epitope antigens compared to epitope-specific maps.
Eventually, this project morphed into the ferret-human distance comparison, but this code could still be used for antigenic cartography.
- R - scripts containing useful functions and code.
- experiments - R notebooks designed to test code and run simulation studies. Notes within each notebook explain the purpose and approach.
- stan - .stan models used to estimate antigenic maps using HMC.
- tests-and-development - R notebooks used to test the inference code and make improvements.
- This Asana project contains some relevant tasks and notes.
- This lab meeting presentation and check-in slides outline the approach and preliminary results.
- This writeup describes the model I was using to generate synthetic titers to multi-epitope antigens.