%% Capillairy Loss Modeling based on experiemntal data
% Written by Patrick Reeson
% Brown Lab, Division of Medical Sciences
% University of Victoria, Victoria BC Canada

% capillary branch order distrubution and obstruction risk loaded from
% excel file
% risk of pruning is 0.3

%% Import data from spreadsheet
% Script for importing data from the following spreadsheet:
%
%    Workbook: C:\Users\P.Reeson\Documents\MATLAB\mod_data.xlsx
%    Worksheet: Sheet1
%
% To extend the code for use with different selected data or a different
% spreadsheet, generate a function instead of a script.

% Auto-generated by MATLAB on 2017/07/06 17:30:37
% clear variables
clear
%% Import the data
%% CD to open excel files with branch order and risk distrubutions
%% File is named mod_data
% mod_data (risk distrubuted across all branch orders based on obstruction distrubution)
moddata = xlsread('C:\Users\P.Reeson\Documents\MATLAB\mod_data.xlsx','Sheet1');

%% Clear temporary variables
clearvars raw;
%% moddata is excel sheet with obstruction risk for each branch order (min 1, max 16 depending on how risk is split) in order
moddata;

% List of all Branch Orders considered
brancho = moddata(:,1);
% get number of Branch Order Bins
nbin = brancho(end);
% Frequency of each Branching Order bin (based on real in vivo data, either
% for each bin or pooled if risk was also pooled (based on moddata file)
binfrq = moddata(:,2);
% Risk of Obstruction for each bin based on pooling stratagey, sum always
% adds up to the experiemntally observed risk for all vessels 
obstrisk = moddata(:,3);
% Starting number of vessels for theoretical 100k cappilairies based on
% experimentally observed distrubution of branch orders and stratagey for
% pooling risk 
startves = moddata(:,4);
% Prune risk is 30% of all obstructed
prunerisk = obstrisk .*0.30;
% Start matrix that will be the model, each row is a branch order's #
% vessels, eacg column is next iteration of 2 hour cycle of obstruction and
% pruning
vmodel = [brancho,startves];
% number of cycles to run
run = 50000;

% Run loop for each 2 hour window pruning iteration
i = 3;
k = 3;
% start with starting distrubution of vessels across branching orders
vmodel(:,1) = startves;
% newves is new # of vessels for each branch order after 1 2 hour window
newves = startves;
for i = 3:run-2
    iprune = newves .* prunerisk;
    newves = newves - iprune;
    vmodel(:,k) = newves;
        i = i+1;
        k = k+1;
end 


% tworun is the total number of hours passed, ie number of 2 hours cycles
tworun = run*2;
xaxis = linspace(2,tworun,run-2);
vtime = transpose(xaxis);
% vsum is total sum of vessels at any time
vsum = sum(vmodel,1);
% vnorm is the normalized number of vessels for each branch order,
% normalized to vsum
vnorm = vsum./100000;
vnormt = vnorm.';
%tvsum is vsum transposed for graph
tvsum = vsum.';
vtime_2 = horzcat(vtime,tvsum);

% Generate each bin as fraction of total vessels for each time
q = 1;
z = 1;

for q = 1:run-2
    for z = 1:nbin
    binf(z,q) = vmodel(z,q)/vsum(q);
    z=z+1;
    end
    q=q+1;
end

figure
% Create xlabel
xlabel('Time (hours)');


set(0,'defaultlinelinewidth',2);
hold on


ax1 = subplot(1,3,1);
plot(ax1,vtime,vnormt)
ax2 = subplot(1,3,2);
p = 1;

ax3 = subplot(1,3,3);
r = 1;
for r = 1:nbin
    plot(ax3,vtime, binf(r,:))
    hold on
end
hold on

% select risk is from BO distrubutions based on gettin x number of 
% obstructions OVERALL if only at select risk vessels obstructed 

u = 1;
a = 1;
vsummat = vsum;
% Concat vert vsum to get a nbinXrun-2 matrix
for a = 1:nbin-1
    vsummat = vertcat(vsummat,vsum);
    a=a+1;
end


for u = 1:run-2
 % number of at risk vessels
        riskfinal = obstrisk.*vmodel(:,u);
        % normalized to  number of vessels in the branch order
        risknorm = riskfinal/vsummat(:,u);
        % sum of all  risk across branch orders 
        sumriskfinal(:,u) = sum(risknorm,1);
       
 
    u = u+1;
end
% starting numbers of at risk population, ie zero is time zero
zerorisk = sumriskfinal(1,1);
% Normalized sumrisk final 
nsrf = sumriskfinal./ zerorisk;
 ax4 = subplot(1,3,2);
 plot(ax4, vtime, nsrf(1,:));
 
set(ax1,'FontName','Calibri','FontSize',14,'FontWeight','bold',...
    'LineWidth',2,'XColor',[0 0 0],'YColor',[0 0 0],'ZColor',[0 0 0]);
% Set the remaining axes properties
set(ax3,'FontName','Calibri','FontSize',14,'FontWeight','bold',...
    'LineWidth',2,'XColor',[0 0 0],'YColor',[0 0 0],'ZColor',[0 0 0]);
 
 set(ax4,'FontName','Calibri','FontSize',14,'FontWeight','bold',...
    'LineWidth',2,'XColor',[0 0 0],'YColor',[0 0 0],'ZColor',[0 0 0]);
% Create legend
legend1 = legend(ax4,'show');
set(legend1,...
    'Position',[0.906051734112396 0.140358770176528 0.0739385045595783 0.795792056634875]);
title(legend1,'Branch order');