Transmission Inference

This repository contains a series of healthcare-associated tranmission tools currently under development.

1. Inference code - C++ source code can found in src
2. Simulation code - R and Python code for simulating transmission events and associated epidemiological and genetic data
3. Reporting - R scripts for checking inferrence outputs against simulations, and visualisation scripts

For further details please contact david.eyre@ndm.ox.ac.uk

---

Running the code

Input files

All times are numbered for input files starting from 1. Assumes only single infection per patient. Assumes only single sample per patient.

1. input/patientLog.csv : file with headers -
   patient_id [string] any patient identfier
   t_inf [int] infection time if know (only used for testing)
   source [string] infection source patient identifier (only used for testing), must be "-1" for backrgound source types and starting the simulation infected
   source_type [int] infection source type (only used for testing: BGROUND_HOSP = 0, WARD = 1, HOSP = 2, BGROUND_COMM = 3, START_POS = 4, SPORE = 5)
   t_sample [int] sample time t_recover [int] recovery time if known (only used for testing)

   where testing values are not known enter "NA" in the cell

2. input/wardLog.csv : file with headers - patient_id [string] any patient identfier
   ward [string] any ward identfier
   hospital [string] any hospital identfier
   t_admit [int] ward admission time
   t_discharge [int] ward discharge time

3. input/simDistancesSNPs.txt - matrix of SNP distances, seperated by spaces, with column and row labels using patient ids, column header has no leading space

Example input files are provided in the example_input folder