Download introduction - Gerstein Lab Publications

Researches have sought to understand and determine a universal molecular phylogeny,
also known as the tree of life. Different sequences of proteins or ribosomes have been
used as a basis of comparison. Currently, the most widely accepted grouping is based on
sequence similarity of small subunit ribosomal RNA (Woese 1987; Woese et al., 1990).
This method uses important and highly conserved genes as the basis of phylogeny which
has complex interactions with many other RNA and proteins. However, the ribosomal
RNA tree is under much question and scrutiny due to problems such as long-branch
attraction, unresolved tree differences, lack of incorporation of other genomic data,
among-site rate variation, mutational saturation, and most recently, lateral gene transfer
(LGT).







Long branch attraction is present when there are differing rates of sequences
change among different sites in a gene, resulting in the clustering of organisms
with higher rates of sequence change.
Trees build on RNA polymerase, which are just as fundamental and complex,
have produced trees that differ from the rRNA phylogeny. One explanation is
secondary deletion events.
Since the small subunit ribosomal RNA corresponds to less than 0.2% of the
genomic material in most microorganisms (i.e., ~1.8 kb for the rRNA in genomes
~1Mb and up), focusing exclusively on it ignores the bulk of the genetic
information in constructing phylogenetic trees.
Different rates of evolution between different genes have resulting in differing
gene phylogenies differ depending on the gene chosen. Organismal phylogeny is
no longer based on just one gene but as a synthesis of the differing gene
phylogenies.
Mutational saturation can greatly affect the deeper branches of the tree as the
possible places of mutation have been exhausted.
“Mutational saturated sequences are maximally diverged, so that further changes
are as likely to make them more similar as they are to make them more different,
and tree topology is based on noise.”
Genes have been shown to be “transferred” from one organism from another,
meaning that when a gene is present in a organism, it is not necessary from its
ancestor. Some researchers have proposed that the rRNA itself can be
transferred, with evidence such as in vitra replacement of the E. Coli SSU rRNA
to or the SSU rRNA hetereogeneity among the Streptomyces.
As a result, other approaches have been used for phylogenetic analysis. Some
researchers still based their methods on sequences, such as ones of ATPases, elongation
factors, aminoacyl tRNA synthase, and beta-tubulin; some of new trees seem more
representative based on macroscopic properties. Other researchers have taken advantage
of the availability of completely sequenced genomes to determine the phylogeny. Gupta
used insertions and deletions, which he called “indels,” along with the structure of the
cell membrane for tree reconstruction. Gogarten suggests building trees using events of
LGT as characters.
The sequencing of whole genomes of microbial organisms allows us to reassess how we
place organisms into groups and relate them to each other in phylogenetic trees. In our
survey, we build trees based on gene function denoted by the COGs database also on
three-dimensional structure classification suggested by SCOP.
As schematized in Table 1, we built trees considering progressively more and more of the
information in a genome and compare them with the traditionally proposed phylogeny.
We are proposing a number of novel trees based on the overall genomic occurrence of
two specific features - folds and orthologs. We call these "genomic trees" or "wholegenome trees." Our aim is to assess how these compare to phylogenetic trees constructed
on other levels.
distance more robust
all trees suffer from long branch
ribosomal distance and parsimony
goal write 2 to three paragraphs robosomal stil
h transfer rates of evolution
moves eu and pro
controversial
googarten general paper
doolittle specific exampoe
edlind specific examples
same thing
find ref describe current cintrivery in whole genome
phylogeny
2 ribosomal problems
2 genome comparison even more unsettled
long branch attraction
clippings
online gavin
* http://bioinfo.mbb.yale.edu/~jimmyln/clippings/Pennisi-Uproot-tree-of-life.pdf
"Comparisons of the genomes then available not only didn't clarify the
picture of ho life's major groupings evolved, they confused it.
"So confusing that some biologist are ready to replace what has become the
standard history with something new."
the new genomes are not adding details to the traditional tree but
challenging the tree altogether
genes unreliable due to horizontal gene transfer
perhaps not even use trees anymore because of lateral transfer
Deinococcus radiodurnas contains genes only present in plants (might also
be gene loss from most other microbes)
mycobacterium tuberculosis contains at least eight human genes
Gupta: use small insertions and deletions called "indels" and structure of
cell membrane to build trees --> fundamental split among bacteria
divides archaea between two bacterial branches
Philippe: genes differ in rates of evolution
Robb: suspected wrong placement based on rRNA, placements of
hyperthemophiles towards the bottom might be an artifact
* http://bioinfo.mbb.yale.edu/~jimmylin/clippings/Pennisi-shake-tree-of-life.pdf
[The new genomic data] "threatens to overturn what researchers thought
they already knew about how microbes evolved and gave rise to higher
organisms."
rRNA history: Woese - successful because identify archaea
based only on one genes --> rRNA regarded as tree of life
Reeve: because of different gene evolution, whole genomes produce
different trees
Horizontal Gene transfer: intertwining branches, swapping back and forth
Aaeo --> different phylogenetic placement based on gene
Soll: sum of all these trees make up the organism
Other problematic genomes: Mjan, Bbur, Aful, Tpal etc.
rampant gene swapping
question the very existence of archae
clippings/doolittle-pnas-genealogy-fun.pdf
2.5 billion years for divergence times of bacteria/eukaryote
vs. 3.5 demanded
possible resolution: hasegawa & fitch - covarion model
residues functional constrained at different periods in protein history
horizontal gene transder
"contamination" mitochondria -> has TIM
*** replication, transcription, and translation machinery less subject to
horizontal gene transfer
most eukaryotic nuclear genomes comes from bactiera, should we still
insist in archaeal/eukaryal sisterhood because we think the genes that
show this.. are less subject to horizontal gene transfer? Are they really?
To waht extent is our desire to look at early evolution in terms of
cellular lineages preventing us from seeing that it is about genes and
their premiscious spread across taxonomic boundaries, which then have no
permanent significance?
* phylogenetic analysis of beta-tubulin sequences
rRNA-anomality as a result of secondary deletion events
beta-tubulin based phylogeny
paromomycin sensitivity confirm beta-tubulin but not rRNA
microsporida related to fungi and animals
* phylogenetic classification and the universal tree
genes from multiple sources and lateral gene transfer
1970s Woese suggest SSU rRNA
less reliable of all genes to experience lateral gene tranfer
rRNA and protein based phylogenies show that artifacts related to within-molecule and
between lineage differences in evolutionary rate and mutational saturation can be
misleading about deep branching and the rooting of the tree
genes with different phylogenies
prominent sources of error or uncertainty in establishing patterns:
mutational saturation
long-branch attraction
among-site rate variation
lateral gene transfer
1) not logical to equate gene phylogeny and organism phylogeny
2) unless organisms are construed as either less or more than the sum of their genes,
there is no unique organismal phylogeny
problem with the ery conceptual basis of phylogentic classification
conflict between SSU rRNA and largest subunit of RNA polymerase
in placement of microsporidia
based on elongation factors
(12-14, 19)
gene trasfer seems unlikely to affect genes for replication, transcription, and translation,
especially rRNA genes.
but could be transfered (34)
(32) LGTs of ribosomal protein genes
zuckerkandl and pauling(2) based on orthologous genes
if diff genes give diff trees, no fair way to suppress this disagreement
*New York Times article
* Orthologs, paralogs and genome comparisons
independent gene duplications, parallelism in evolution of independent heritages
with new databank searches for homologous sequences, and increase availability of
reliable three-dimensional structual information
proteins with diverse functions can belong to same superfamily
vertical inheritance not sufficient
horizontal gene transfer (15-17)
wolf (40) studied distribution of protein folds in completely sequenced genomes
ATPases(48), elongation factors(49-51), aminoacyl tRNA synthase(52)
replcation, transcription, and translation machinery (54-55)
other characters (56-59) placement of archae
rooting of tree of life as working hypothesis and not as established fact (60-61)
gene duplicatoion(82,93,85,87)
* Felsenstein, Sys. zool. 27, 401 (178)
long branch attraction occurs when rates of sequence change differ substantially between
taxa. lineages with higher rates of sequence change artifactually associcate with each
other and with out groups, excepts with maximum likelihood methods.