multiple non uniform distributions most common values identification

I need to find the most common values from the 3 main distributions (see figure) from dataInput2.

The code below computes the wrong bins (1-2-3) because the distributions are not normal. Any ideas how I could get the values A-B-C instead? (see figure).....Distributions can change limits, so defining boundaries is not an option

    [counts, edges] = histcounts(dataInput2(:,i),500);
    temp = sort( edges, 'descend' );
    spinPeaks = temp(1:injec);

[CPOR]

[counts, edges] = histcounts(dataInput2,500);
  temp = sort( edges, 'descend' );
  spinPeaks = temp(1:injec);

[CPOR]

above is the actual code that I'm using...

Answer 1

[Image Analyst]

Call findpeaks() on counts().

By the way, why is B not at "2"?

[Image Analyst]

CPOR's "Answer" moved here since it's not an answer to his original question.

Thank you all for your notes... I don't have the signal Processing Toolbox available in order to use findpeaks. Any suggestions how to code it? and yes...the sorting should be on the Counts and not the edges as I mentioned in my original code..... but I still have the same problem: Obtaining 1-2-3 instead of A-B-C

[Image Analyst]

You can just do

[counts, edges] = histcounts(dataInput2,500);
[sortedCounts, sortOrder] = sort(counts, 'descend');
highest3counts = sortedCounts(1:3);
edgesOfHighest3 = edges(sortOrder(1:3));

This will give you 1, 2, and 3 on your graph, not A, B, and C.

[CPOR]

correct...what I need is A-B-C

[Image Analyst]

So buy the Signal Processing Toolbox, or else write findpeaks yourself. Or search the internet for things like https://www.mathworks.com/matlabcentral/fileexchange/25500-peakfinder-x0-sel-thresh-extrema-includeendpoints-interpolate

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2631518/

Also see attached for another one.