apclarkva / paci-populations Goto Github PK

The objective of this code is to develop a monolayer of coupled iPSC Paci 2013 models to simulate the conductance through cardiac tissue. This project is a work-in progress, but the README below explains how to generate a population of models using an LHS method.

There are two .mat files in the data/ folder:

• lhs_parameters_40000.mat — has parameter values for 40,000 potential parameter sets. The variable's size is 8 x 40,000.
• passing_features.mat — has a reference to the parameter set, a flag for whether the parameter set passed, and a flag for whether the parameter set passed each feature. The variable's size is 3,204 x 10.

I realize, I should have made the lhs_parameters_40000.mat file 40,000 x 8, but don't want to mess with that now. I'll circle back and clean things up later.

Anyway, the rows of this array contain values for the following parameters:

1. g_CaL
2. g_K1
3. g_Kr
4. g_Ks
5. g_Na
6. g_PCa
7. g_b_Ca
8. g_b_Na

This contains information about how each parameter set in the 40,000 performed, relative to the Ma features — I can share that stuff with you when I have more time. If you're up to it, you can poke around in the ap_features directory.

Of the 40,000 parameters, only ~3,204 resulted in spontaneous, and AP-like waveforms. The columns of this variable are:

1. parameter index — integer
2. passing AP or not — 1 or 0 (I can't remember if I set the threshold to passing all 7 of the following features, or just 6 — it doesn't matter, because I used column ten to filter for APs that passed 7 vs 6 vs 5...)
3. cycle_length — 1 or 0
4. spontaneous_heart_rate — 1 or 0
5. maximum_diastolic_potential — 1 or 0
6. peak_voltage — 1 or 0
7. action_potential_amplitude — 1 or 0
8. maximum_upstroke_velocity — 1 or 0
9. apd90 — 1 or 0
10. Sum of columns 3 through 9. A passing AP has a value equal to 7.