How can compute feature vector by using GMM and Maha distance

I have trained my GMM by using

    MFCCs(isnan(MFCCs)) = [];
    A = gmdistribution.fit(MFCCs',1);

However, it gives me an error when I want to compute the distant

    MFCCs(isnan(MFCCs)) = 0;
    mahal(A,MFCCs(1,:)')

This error is appeared:

    Error using mahal (line 34)
    Requires the inputs to have the same number of columns.