The "smorfam" column of the output tsv file contains names like "smorfam00946"
How do these families compare to the data in Sberro et al. 2019? Looking at Table S3 from that paper, the first column contains a "family ID" that does not map to the smORFfinder output.
It's also not just the integer order, since the length of the sequences in Table S3 and the hmm file are different.