Interpretation of multicomparison test results of MATLAB to give the letters to samples

I am conducting a one-way ANOVA test on my experimental results. After that, I want to know whether the means of the groups are significantly different from each other. For that I used the Tukey's HSD test under the multicompare test of MATLAB's statistical toolbox. I wanted to know how to interpret the MATLAB results to provide "letters" to each of the groups ? The "letters" (like "a, b, ab" etc.) are very prominently used in research papers to indicate whether there is any significant difference between the means of groups.

It would be a great help if someone guides me in this case. I used the following code;

>> z=[ENFCUE;EBFCUE;DNFCUE;DBFCUE;MFCUE];

>> groups=[ones(size(ENFCUE));2*ones(size(EBFCUE));3*ones(size(DNFCUE));4*ones(size(DBFCUE));5*ones(size(MFCUE))];

>> [p,table,stats]=anova1(z,groups);

>> [c, m, h, nms] = multcompare(stats,'alpha',.05,'ctype','hsd');