Use NCBI site. From aa sequence, you can easily figure out type of enzyme and the conserved site residues. If in doubt, do the homework and look for analogs. Typically, you find matching residues close to the ones identified from aa sequence.

thanks  wieslaw...

You can model the structure using a bioinformatic server, SWISSMODEL,....(if there is a PDB structure with an identity) and after this can also use HotSpotWizard, that it is used for protein enginering.  https://loschmidt.chemi.muni.cz/hotspotwizard/

It is a structure analysis server based on data-mining of reported structures and literature

Thank you Jesus....