I suggest that torsional stresses be reported for beam elements, currently clearly missing from the "worst" stress result combination. Other FEA codes provide this function. Certainly, we aren't considering the "worst" stress if we ignore potententially controlling torsional stresses.