Biology Reference
In-Depth Information
of DNA alignments to detect orthologous genomic regions (synteny
blocks) therefore rapidly deteriorates with the rising level of divergence
between the species being compared.
In practice, DNA-level whole-genome-scale alignments can be per-
formed reasonably well when conservation is such that there are still
relatively long almost-identical substrings, which are therefore unique
and can be used as orthologous anchors, as implemented in CHAOS
58
or
MUMmer.
59
These orthologous anchors can then be used to either guide
local alignment algorithms like BLASTZ,
60
an optimized version of
gapped BLAST, or define synteny blocks, as applied in Cinteny,
61
and
employ global (e.g. LAGAN
62
) or semiglobal alignments (e.g. so-called
“glocal” combinations of global and local methods introduced in
Shuffle-LAGAN
62
). These tools, however, already begin to reach their
limit at the divergence level of, for example, chicken and human or dis-
tant
Drosophila
species, when the orthologous DNA signals are mostly
derived from separate exons and barely distinguishable from random
noise. The delineation of synteny blocks across deeper phylogenies can
therefore benefit from the use of protein translations of predicted genes
to resolve complex orthologous relations. In addition, instead of con-
ventional gap opening and extension penalties, the explicit modeling of
insertions and deletions as proposed in MCALIGN
63
seems to produce
more realistic alignments of noncoding regions.
While many genome arrangement variations are to some degree
tolerated in populations such as segmental duplications, which lead to
copy number variants, or long-scale deletions more frequently associated
with genetic diseases, the most radical change of genome architecture is
arguably that of whole-genome duplications (WGDs). A few WGDs have
been documented in yeast and vertebrate lineages, and have contributed
significantly to the current repertoire of genes and species. Current
understanding of the effects of WGDs predicts a drastic loss of genes fol-
lowing a WGD event by pseudogenization, which subsequently slows
down.
64
The recent comparative analysis of the genome of the unicellu-
lar ciliated eukaryote,
Paramecium
, revealed at least three successive
WGD events, where the most recent duplication was suggested as the
driving force of speciation due to incompatible differential gene losses
that gave rise to a complex of 15 sibling species.
65
Functionally linked
