function xp_comparison_plot_2D(xp, op)
if nargin < 2, op = struct; end
if isempty(op), op = struct; end;
op = struct_addDef(op,'test_handle',@ttest2);
op = struct_addDef(op,'significance',.05);
op = struct_addDef(op,'transpose_flag',0);
op = struct_addDef(op,'flip_axis_flag',0);
op = struct_addDef(op,'xscale','linear');
op = struct_addDef(op,'yscale','linear');
% if nargin < 2, test_handle = []; end
%
% if isempty(test_handle), test_handle = @ttest2; end
%
% if nargin < 3, significance = []; end
%
% if isempty(significance), significance = .05; end
%
% if nargin < 4, transpose_flag = []; end
%
% if isempty(transpose_flag), transpose_flag = 0; end
%
% if nargin < 5, flip_axis_flag = []; end
%
% if isempty(flip_axis_flag), flip_axis_flag = 0; end
%
% if nargin < 6, scale = {}; end
%
% if isempty(scale)
% scale = {'linear', 'linear'};
% end
%
% if isscalar(scale)
% scale_in = scale;
% clear scale
% scale{1:2} = scale;
% end
[xp_dims, xp_sort_index] = sort(size(xp), 2, 'descend');
if xp_dims(1) ~= 2 || xp_dims(2) ~= 1
error('xp_compare_2D can only be used with an xp object having 2 by 1 (or 1 by 2) xp.data_pr.')
end
xp_dim_compared = xp_sort_index(1);
meta = xp.meta;
for d = 1:2
dim_name = ['matrix_dim_' num2str(d)];
if isfield(meta, dim_name)
axis_labels{d} = meta.(dim_name).name;
axis_values{d} = meta.(dim_name).values;
else
axis_labels{d} = dim_name;
axis_values{d} = 1:size(xp.data{1}, d);
end
end
if op.transpose_flag
xp.data = cellfun(@(x) x', xp.data, 'UniformOutput', 0);
axis_labels([1 2]) = axis_labels([2 1]);
axis_values([1 2]) = axis_values([2 1]);
end
if op.flip_axis_flag
indices = cell(1, ndims(xp));
indices(:) = {':'};
indices(xp_dim_compared) = {[2 1]};
xp.data = xp.data(indices{:});
xp.axis(xp_dim_compared).values = xp.axis(xp_dim_compared).values([2 1]);
end
for sample = 1:2
[length_sample(sample), n_sample(sample)] = size(xp.data_pr{sample});
end
%% Compare columns across rows.
no_tests = min(length_sample);
p_values = nan(no_tests, 2);
for t = 1:no_tests
[~, p_values(t, 1)] = feval(op.test_handle, xp.data_pr{1}(t, :)', xp.data_pr{2}(t, :)', 'tail', 'left');
[~, p_values(t, 2)] = feval(op.test_handle, xp.data_pr{1}(t, :)', xp.data_pr{2}(t, :)', 'tail', 'right');
end
test = p_values < op.significance/2;
%% Find mean and s.e.
plot_length = max(length_sample);
[sample_mean, sample_se] = deal(nan(plot_length, 2));
for sample = 1:2
sample_mean(1:length_sample(sample), sample) = nanmean(xp.data_pr{sample}, 2);
sample_se(1:length_sample(sample), sample) = nanstd(xp.data_pr{sample}, [], 2)/sqrt(n_sample(sample));
end
%% Plot w/ stars for signifcance.
if length(axis_values{1}) < plot_length, axis_values{1}((end + 1):plot_length) = nan; end
boundedline(axis_values{1}, sample_mean, prep_for_boundedline(norminv(op.significance/2)*sample_se))
set(gca, 'XTickLabelMode', 'auto', 'YTickLabelMode', 'auto')
axis tight, box off
add_stars(gca, axis_values{1}(1:no_tests), test, [1 0], [1 0 0; 1 .5 0])
%% Make legend.
mylegend = cell(1, 2);
for sample = 1:2
if isnumeric(xp.axis(xp_dim_compared).values)
mylegend{sample} = sprintf('%s = %g', xp.axis(xp_dim_compared).name, xp.axis(xp_dim_compared).values(sample));
elseif iscellstr(xp.axis(xp_dim_compared).values)
mylegend{sample} = sprintf('%s = %s', xp.axis(xp_dim_compared).name, xp.axis(xp_dim_compared).values{sample});
end
end
legend(mylegend)
set(gca, 'XScale', op.xscale)
set(gca, 'YScale', op.yscale)
end