taxa identification using signature vs alignment to refseq

Question

[bitpir]

Has anyone tried and compared the taxa identified using both signatures and reference sequences? I seem to identify much more taxa from the refseq with little overlap with what i identified from signatures. In this case, which results should i trust more?

[GenoMax]

Which do you think is more reliable? A likely (incomplete) signature databases versus experimentally/manually validated RefSeq sequence.