This is a Research Dataset associated with the journal article

Mortensen, P., Gao, H., Smith, G., Simitev, R., (2020)
Action potential propagation and block in a model of atrial tissue
with myocyte-fibroblast coupling.
Mathematical Medicine and Biology
https://doi.org/10.1093/imammb/dqaa014

1) Data formats and software for analysis

The data is either in ascii text format or in the format of MATLAB
".mat" files. MATLAB scripts for analysis of .mat files are provided
and have file extensions ".m".

A bibtex reference for MATLAB is
@manual{MATLAB:2017b,
address = {Natick, Massachusetts},
organization = {The Mathworks, Inc.},
title = {{MATLAB version 9.3.0.713579 (R2017b)}},
year = {2017}
}


2) Description of datasets

The data for Figures 03, 04, 06a, 08a, 11, 12, 13, 14 is contained within
PaperFigures.m, which will plot the data and the data within the .mat files.

To load the .mat files in Matlab using the function load(), e.g. load('Ana0.mat')

Ana0.mat, Ana1.mat, ..., Ana9.mat Contain the data of the analytical curves for N=0,1,..,9 
and Ana10.mat contains the data close to the threshold (i.e. when V_alpha=-66.67mV), as seen in Fig10,
The .mat files contains for types of data, denoted by the prefix of the array name, 
the V_alpha (Va*), V^omega (V*), the wave speed (c*) and the gating variable j (j*)
If there is no suffix the suffix *up.mat the data is plotted together to create a curve as outputted from the Fortran code.
If the suffixes are B and R, these represent the same curve, but separated about the minimum value of j, as seen in Fig 10. 
R being the unstable half plotted in Red.
B being the stable half plotted in Blue.
I is a number indicating the index in the for arrays that aligns with the minimum value of j, in the arrays without suffixes (V, Va, c, and j).
Note that

APD30a.mat, APD30b.mat APD90a.mat, APD90b.mat, APDRa.mat, APDRb.mat contains the data seen in Fig04g,h,i. 
Note each data set has two files (denoted with *a.mat, and *b.mat) for each of the two branches of the alternans.
In each .mat contains a cell of 11 arrays. The 11 arrays correspond to a different number of fibroblasts per myocyte. APD*{1} corresponds to N=0,
APD*{2} corresponds to N=1, and so on, until APD*{11} is N=10.
Each array contains the respective APD value over a range of BCLs (300, 400, ...,1000ms). An appropriate array for the BCLs can be found in BCLS.mat. 

The other .mat files (for Figs 02, 05, 06b, 07, and 08b) contain an array called DataMatrix. 
Each column is the myocyte potential (in mV) over a linear cross-section of the mesh at a single time step.
The first column is the first recorded time step, the time progresses sequentially along the columns, in time steps of 2.5ms.
The number of rows is is dependent on the spatial step (dx) used, but step size is the same throughout across cross-section.
The number of columns is dependent on time span and which time points were recorded.

Test12Vm.mat and Test12Vf.mat contain the data of myocyte and fibroblast AP propagation as seen in Fig2.
The Distance 150mm (1500 rows), the time span is 0ms to 1000ms (400 columns).

Fib15Block05.mat, Fib15Block10.mat, and Fib18Block05.mat contain the example cases of Case 2, as seen in Fig05.
The cross-section length is 50mm with dx=0.1mm(500 rows), the time span is 2.5-2000ms with a time step of 2.5ms(800 columns).

Fib15Block05Ca.mat, contains the Ca profiles seen in Fig06b
The cross-section length is 50mm with dx=0.1mm(500 rows), the time span is 2.5-2000ms with a time step of 2.5ms (800 columns).

Y004Fib15.mat, Y006Fib15.mat and Y006Fib18.mat contain the data seen in Fig07.
The cross-section length is 50mm with dx=0.05 (1000 rows), the time span is 1000-1500ms with a time step of 10ms (51 columns).

Fib15Y006Ca.mat, contains the Ca profiles seen in Fig08b.
The cross-section length is 50mm with dx=0.05 (1000 rows), the time span is 1000-1500ms with a time step of 10ms (51 columns).

jminE and jminEmore are data files for the minimum value of j, as seen in fig10 and Fig 13. Can be loaded in Matlab with load('jminE') and load('jminEmore')

There is also the Fortran Code, jc.FD.F, to solve the reduced model.