I think it is not required to merge all of the predicted structures, rather use only one software that is frequently cited in published articles like GOR4 (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_gor4.html) and psipred server (http://bioinf.cs.ucl.ac.uk/index.php?id=779), that would be enough, PLEASE correct me if I'm wrong