Intersection of yline and x,y categorical data

I will not show the code which produces the data in the attached figure as I feel it is rendundant with my current issue.

In the figure attached, I have a series of points for which I have joined them by a line.

I also have a yline (50%) at which passes the previous plot line, once, around -7.5M. However, I want to have an exact value for this and to make the process automatic, rather than manual.

I have tried using various functions like interX and so on from file exchange but to no avail.

The data, and current plotting code is below:

Note - I did remove the first point of the data (ND) and then converted the xdata into numerical data, on the log scale but still had issues. Obviously, I removed the associated point from the yData vector.

I understand that this is probably the better way to do it, and then just plot the categorical case on top of the numerical case, like my code does.

% the original categorical x data
xS=categorical(["CTR" "-12" "-11" "-10" "-9" "-8" "-7"]);
xS = reordercats(xS,{'CTR' '-12' '-11' '-10' '-9' '-8' '-7'});

%Numerical version of above, removing the CTR entry (re-added back later)
cd = [1e-12 1e-11 1e-10 1e-9 1e-8 1e-7];

xS2 = log10(cd);

%The two different y data sets for each x data above
yS = [1 0.9995 0.9978 0.9815 0.8571 0.5576 0.4465]*100;
yS2 = [0.9995 0.9978 0.9815 0.8571 0.5576 0.4465]*100;

fig = figure(2);
%Actual plots
plot(xS,yS, '-o','Color','[0, 0.4470, 0.7410]',...
    'LineWidth',3.5,'MarkerSize',20,...
    'MarkerFaceColor','[0, 0.4470, 0.7410]','MarkerEdgeColor','k',...
    'Linewidth',3); 
hold on
yline(50, 'r--','Linewidth',3.5);
% Second to calc Ic50 value - uses numeric data
plot(xS2,yS2, '-o','Color','[0, 0.4470, 0.7410]',...
    'LineWidth',3.5,'MarkerSize',20,...
    'MarkerFaceColor','[0, 0.4470, 0.7410]','MarkerEdgeColor','k',...
    'Linewidth',3);

xlabel('Dose [logM]'), ylabel('% Control')
leg = legend({'Cell Number','IC_5_0'});
leg.LineWidth=2.5;
leg.FontSize=28;
leg.NumColumns=1;
leg.LineWidth=2;

% Other plot stuff
box on
ax = gca;
set(gca,'linewidth', 2.5,'fontsize',28,'fontname','Helectiva') % Sets the width of the axis lines, font size, font
set(gca,'TickDir','out')
x0 = 50;
y0 = 50;
width=800;
height=600;
set(gcf,'position',[x0,y0,width,height])

Answer 1

[Star Strider]

Since the ‘y’ values are monotonically decreasing, using interp1 is straightforward with the second vector, however there is no way to work with the categorical variables to interpolate them, so I just went with the indices instead.—

% the original categorical x data
xS=categorical(["CTR" "-12" "-11" "-10" "-9" "-8" "-7"]);
xS = reordercats(xS,{'CTR' '-12' '-11' '-10' '-9' '-8' '-7'});

%Numerical version of above, removing the CTR entry (re-added back later)
cd = [1e-12 1e-11 1e-10 1e-9 1e-8 1e-7];

xS2 = log10(cd);

%The two different y data sets for each x data above
yS = [1 0.9995 0.9978 0.9815 0.8571 0.5576 0.4465]*100;
yS2 = [0.9995 0.9978 0.9815 0.8571 0.5576 0.4465]*100;

xSint = interp1(yS(2:end),(-12:-7),50)                          % 'xS' Must Be Numeric

xSint = -7.4815

xS2int = interp1(yS2,xS2,50)                                    % 'xS2' 'yS2=50' Intercept

xS2int = -7.4815

fig = figure(2);
%Actual plots
plot(xS,yS, '-o','Color','[0, 0.4470, 0.7410]',...
    'LineWidth',3.5,'MarkerSize',20,...
    'MarkerFaceColor','[0, 0.4470, 0.7410]','MarkerEdgeColor','k',...
    'Linewidth',3); 
hold on
yline(50, 'r--','Linewidth',3.5);
% Second to calc Ic50 value - uses numeric data
plot(xS2,yS2, '-o','Color','[0, 0.4470, 0.7410]',...
    'LineWidth',3.5,'MarkerSize',20,...
    'MarkerFaceColor','[0, 0.4470, 0.7410]','MarkerEdgeColor','k',...
    'Linewidth',3);

xlabel('Dose [logM]'), ylabel('% Control')
leg = legend({'Cell Number','IC_5_0'});
leg.LineWidth=2.5;
leg.FontSize=28;
leg.NumColumns=1;
leg.LineWidth=2;

% Other plot stuff
box on
ax = gca;
set(gca,'linewidth', 2.5,'fontsize',28,'fontname','Helectiva') % Sets the width of the axis lines, font size, font
set(gca,'TickDir','out')
x0 = 50;
y0 = 50;
width=800;
height=600;
set(gcf,'position',[x0,y0,width,height])

Create a numeric vector for the categorical arrays to work with them, if my approach is incorrect.

.

[Alistair McQueen]

Okay, I guess it's been a long week... Was not expecting it to pop out as simply as that given the other examples I have been looking at today.

Thanks very much!! That is the correct value. So, I just need to plot it now which is easy. The CTR value is redundant when figuring out this intercept as the y data is a function of that value anyway, so just using the numerical data is fine.

Enjoy your weekend!!

[Star Strider]

As always, my pleasure!

Your data are such that the interpolation is straightforward. Other times, it requires isolating the region of one or more of the intersections, then interpolating.

You, too!

Enjoy Samhain!

.