Dear scientists/experts.

I have two sequenced isolates (16s rDNA) recently searched using BLAST-N, the results showed that the first has a high query coverage (95-100) and low similarity (92.65-94.15%), while the second has a query coverage of 99% and higher similarity (96.46-97.11%). In this case, is the query coverage more important than the similarity, or the other way around? If the similarity was more important, what similarity percentages are acceptable enough?

Thank you in advance