In a side conversation, Chris Somerville and I have been discussing the need
for better transgene nomenclature. Chris suggested that I have a go at
listing some proposals for discussion here. Here goes:
1) Transgene loci are identified as such by the prefix "t" (by analogy to the
bacterial plasmid "p" designation).
Example: a chalcone synthase transgene would be t-chs (or tchs or t.chs?)
2) Because transgenes can insert at many loci, each is distinguished
numerically as t-chs1 through t-chs1000 and beyond.
3) Because one can construct an infinite number of chs transgene variations
(different promoters, different linked drug markers, etc), there has to be a
mechanism to avoid two labs using the same designation. Thus, symbols and
blocks of numbers would be registered with the database (i.e., applied for
and assigned). Thus, Fred Ausebel might be assigned t-chs1-100 and Joe
Schmo, t-chs101-110. (Note: This is exactly what was adopted by bacterial
geneticists in 1976 for plasmid nomenclature. Unfortunately, e-mail was not
available to "spread the word" and to make the process of registering symbols
painless. As the field exploded, the system was overrun; hence many
constructs were called things such as, p12X/lac-4bnptII#1, and that sort of
nonsense (which looks a lot like many transgene designations in plants).)
4) Transgene loci should be assigned symbols only when genetically
characterized as single Mendelian loci.
5) Transgene alleles can be designated by the same rules as for endogenous
genes, e.g., t-chs210-4.
6) Symbols can be any three letter combination followed by one or more