Gene Bank and EMBL are both indiscriminate main databases. As long as it is obtained through experiments, no matter what sequence, even incomplete sequences can be uploaded, and their data may be copied. If someone specializes in identifying bacteria, it is necessary to use the officially recognized 16srDNA sequence. For the convenience of research, the data of the recognized standard 16srDNA sequence of various bacteria in these primary databases are sorted out, and a database, the so-called secondary database, is reconstructed. If you don't build it, you can directly use the primary database for blast, and you will get a lot of unrecognizable or even incomplete sequences. It is very troublesome to manually look at them one by one to find out the recognized standard sequences. In reality, the example I gave was EzTaxon in South Korea.