William R. Pearson
wrp at dayhoff.med.Virginia.EDU
Tue May 10 09:57:57 EST 1994
In article <9405051451.AA20816 at thalamus.wustl.edu>,
Yamagata =?ISO-2022-JP?B?GyRCOzM3QRsoQg==?= <yamagatm at THALAMUS.WUSTL.EDU> wrote:
>How do you define a homolog, ortholog, or paralog? Family, superfamily,
>subfamily, cognate, counterpart, related, homologous, --Do you use other
>words for describing relationships among homologous genes? And, how would
>you use these words? Is there good references for how we can accurately
>distinguish these words? I know there is no simple answer for this. But,
>I would like to know how molecular evolutionalists are using these words.
>Thanks in advance.
Homolog (homologous) - shares a common ancestor
ortholog - differences between two "orthologous" proteins reflect
a speciation event - mouse alpha-hemoglobin and human
alpha-hemoglobin differ because of the divergence
of the lines leading to mice and humands
paralog - differences reflect gene duplication events, e.g.
human beta-hemoglobin and mouse alpha-hemoglobin.
An evolutionary tree based on chimpanzee alpha-globin,
human beta-globin, and mouse beta-globin would put
humans much closer to mice than chimps.
This definition is available in any textbook on molecular evolution.
The superfamily/family/subfamily hierarchy was defined by Dayhoff.
Typically members of a subfamily are more than 80% identical, family >
50%, and superfamily < 50%. (Meth. Enz. 1990 183:37)
More information about the Mol-evol