When you type your name into the box, the program reads the letters of your name as if they were the single-letter codes for amino acids. Since there are only 20 amino acids, if you have a B, J, O, U, X, or Z in your name the program reads it as “X” which just means any amino acid could go in that spot.
The amino acids are then translated back into one of the possible three-letter DNA codes for each amino acid, and that DNA sequence is searched against the genome databases for the protein that has the closest match to your name.
My name becomes TGCCACAGAATCAGCACCATCAACGCCGCCGGCGCCCCCGCCAAGATCAGC
in DNA, and the closest protein match is Phosrestin from fruit flies (apparently they allow a lot of gaps in the search), a protein involved in the fly photoreceptor.
Synthetic biologists like Craig Venter sometimes like to code their own “watermark” into artificially synthesized DNA sequences as a way to sign their work, so it’s fun to be able to turn it around and search for words already there (btw, Craig Venter’s closest protein name match is Fructose-bisphosphate aldolase 1 from the garden pea).
What’s your protein name?