Information Technology Reference
In-Depth Information
10
9
8
7
6
t
I
5
4
3
2
1
0
0
200
400
600
800
1000
1200
1400
E. coli genes
Fig. 15.9. Mean translation time
I
for 1530 Escherichia coli coding regions calculated using
data on tRNA abundances published by Ikemura [46]. The sequences are ordered according to
the values of
I
. The lower line shows
I
for the correct reading frame. The upper lines show the
value of
I
for the +1 and +2 reading frames. The open squares correspond to ribosomal genes,
the lled triangles correspond to genes carried by plasmids.
protein to its tri-dimensional shape is still unsolved. Even more dicult is the
problem of computing the eects of this modied shape in the chemical network of
a cell. Therefore, the only landscape that is considered is the neutral (at) one.
One of the few cases in which it is possible to guess the consequences of mutations
is the case of synonymous substitutions in bacteria [45].
The genetic code (correspondence between codons and amino acids) is highly
redundant: there are 64 dierent codons, and only 20 amino acids, plus a stop
command that signals the end of the protein chain. This redundancy occurs in two
ways: the third codon is often uninuenced for the binding of the tRNA, and many
dierent tRNA (cognate to dierent codons) correspond to the same amino acid
(synonymous tRNA). Therefore, all these substitutions are neutral with respect to
the protein produced. Synonymous tRNA occurs in dierent abundances.
The time needed for a ribosome to attach a new amino acid to the growing
chain depends mainly on the waiting time for the right charged tRNA to arrive.
This time is proportional to the abundance of tRNAs, and supposing that all of them
are always charged, it is proportional to the copies of the tRNAs in the genome.
Therefore, even if synonymous substitutions are neutral for the functionality of
proteins, they imply dierent time intervals for their production.
Bacteria in an abundant medium grow exponentially, and a small dierence in
the time needed for a duplication can make a big dierence in tness. Proteins are
needed in various abundances for duplications: structural ones, like ribosomal pro-
teins, are essential (and consequently they are well optimized and very conserved),
while those carried by plasmids, being transmittable among dierent strains, are
presumably not very optimized. This assumption is conrmed by Fig. 15.9, in
which it is reported the translation time
1
(computed using the abundances of
