# What is the meaning of scores when comparing reaction classes in the KEGG database

With respect to related R class in KEGG website:
http://www.genome.jp/tools-bin/relclass?RC00064

What is the biological or statistical relevance of the score? How is it computed? And does it signify the bonding strength of how probably it is for a pair of compounds to be an R pair?
Note: This may be a silly question to all chemists, but not for me, i am a computer science student , and my research field in bioinformatics and so i have to study both biology and bit of chemistry, so please just dont close the question or be harsh on me, I have very little knowledge in chemistry and also googling doesn't help much. I asked this question in Biology forum, they closed the question and asked me to ask it here

• This question could also profit from more context and an actual title. – Martin - マーチン May 19 '15 at 4:31
• genome.jp >> KEGG >> KEGG Overview >> R number – John Snow May 19 '15 at 5:42

RC00064 apparently is an N-acylation or the formation of a peptide bond. RC00197 with a score of 1.0 is a N-formylation and considered very similar.
Change the cutoff to a lower value, such as 0.4, and compare the results at the end of the table with RC00064. You'll find dissimilar reaction classes like RC02518, the oxidation of a selenol to a selenic acid.