Re: Fw: Most popular/common dinosaur misconceptions

[David Marjanovic <david.marjanovic@gmx.at>]

> > Why? The tRNAs will still not bind where they should, and neither will the
> > ribosome release factor that terminates translation. Lateral gene transfer
> > can only work if the genetic codes are the same.
>
> Translation of the whole gene can only occur if the genetic codes are the
> same.  There may be transcription and translation of a recently arrived 
> gene
> (i.e., by lateral transfer), but the polypeptide product may be truncated.
> It depends when the first TGA comes up.  Tryptophan is one of the less
> common amino acids.

I know. There are even a few small proteins without tryptophan. But I was 
talking about the general case -- two completely independent genetic codes. 
This would lead to a lot more problems.

Even the vertebrate mitochondrial code has more than this single difference: 
an isoleucin codon is read as methionin(e?), and two arginine codons have 
become stop codons. So, our cytoplasmatic ribosomes would merrily ignore two 
out of four stop codons and just keep on translating, producing a completely 
dysfunctional protein.

Now, the vertebrate mitochondrial code has obviously evolved from the 
"universal" code. Imagine two codes that have evolved independently. Will a 
gene that belongs to one code look different than a random sequence of 
nucleotides to the other code?

And even this preassumes that the exact same amino acids and bases are 
used...

> (Among bacteria, mycoplasmas also have their own genetic code.)

Sixteen codes are known: 
http://www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi?mode=c#SG1 The 
differences are always few in number, and involve stop codons or amino acids 
that are chemically similar to each other, arguing strongly for monophyly of 
all known codes, as do features like the fact that all animal mitochondria 
seem to read UGA as tryptophan instead of as a stop codon.