COMPUTER CODES RELATED TO

Carruthers J, Lythe G, Lopez-Garcia M, Gillard J, Laws TR, Lukaszewski R and Molina-Paris C (2020) Stochastic dynamics of Francisella tularensis infection and replication. PLoS 
Computational Biology.

Author: Jonathan Carruthers (Emergency Response Department, Public Health England, Porton Down, Salisbury, SP4 0JG)
Date: 25 March 2020
Correspondence for questions: jonty.carruthers@btinternet.com

These computer codes can be used to reproduce results in "Carruthers J, Lythe G, Lopez-Garcia M, Gillard J, Laws TR, Lukaszewski R and Molina-Paris C (2020) Stochastic dynamics of 
Francisella tularensis infection and replication. PLoS Computational Biology ()

Programming Language: Python
Python version used: Python 3.6.10
Operating system: Windows 10

Description of .py files

"FT ABC.py": Applies an Approximate Bayesian Computation (ABC) rejection sampling algorithm to infer parameters of the agent-based model (ABM) from an in vivo study of Francisella 
	     tularensis infection in mice. For efficiency, the code makes use of the cohort analysis to approximate the ABM. Samples from the approximate posterior distributions for each
	     parameter are written to a .txt file, along with the corresponding value of the distance function. These samples can be used to construct pointwise predictions (Fig 9, 12) and
	     posterior histograms (Fig 10, 11).

"FT Agent-based model.py": Simulates realisations of the ABM used to describe the early stages of Francisella tularensis infection in multiple organs of a mouse. For a specified
			   parametrisation, output is written to four separate files:
				i) fttauout.dat - for the lung, liver, spleen, kidney and MLN, records the number of extracellular bacteria, phagosomal bacteria, cytosolic bacteria,
						  uninfected macrophages, suppressed macrophages (infected and uninfected), resting macrophages and activated macrophages, as well as the
						  concentration of IFN-gamma and TGF-beta.
				ii) cohort.dat - for the lung, liver, spleen, kidney and MLN, records the number of phagosomal and cytosolic bacteria in the first five cohorts
				iii) cohortmac.dat - for the lung, liver, spleen, kidney and MLN, records the number of cohort n macrophages for n=1,2 (the cohort number of a macrophage
						     is equal to the cohort number of the bacterium that first infects it).
				iv) rupture_times.dat - records the times at which macrophages rupture, the cohort number of the rupturing macrophage, how many bacteria are released, and
							the organ within which it occurs. 

"FT Cohort analysis.py": Uses the cohort analysis to approximate the number of bacteria in macrophage phagosomes and cytosols within multiple organs of a mouse, reproducing the predictions
			 provided in Fig 7.