Currently writeResults creates output that is structured as results/analysis-name/resultsmatrix, but if the user wants to put this back into ModelArray for subsequent analysis (e.g., in the case of harmonized data) this requires substantial reformatting of the H5. It would be awesome if writeResults could have an argument (like -analyzable?) that would format the output into an H5 that could then be read back into ModelArray for subsequent analysis!