Francisco - The highly successful Deep Green project to construct
a "tree of life" for the green plants has ended, but it has
seeded new projects to strengthen the branches and root the
tree more firmly in new genetic and fossil data.
Among these projects is "Deep Gene," headed by University
of California, Berkeley, botanist Brent D. Mishler, and
"Deep Time," headed by Doug Soltis of the University of
Florida. The National Science Foundation (NSF) has agreed
to fund both projects with $500,000 each over the next five
The success of Deep Green also has emboldened NSF to float
the idea of a much larger project - generating the definitive
tree of life for everything, from bacteria to bats, fungi
to flowering plants. NSF director Rita Colwell calls Deep
Green one of the best investments the foundation has made,
Mishler and four colleagues will brief reporters Feb. 16
at 11 a.m. about the accomplishments of Deep Green and its
proposed offshoots. Mishler, a spokesman for Deep Green,
is director of UC Berkeley's University and Jepson Herbaria
and a professor of integrative biology in the College of
Letters & Science.
Deep Green has contributed to more than 100 research papers,
Mishler said, the latest of which, in the Feb. 1 issue of
Nature, nailed down the sister group of the seed plants.
The work, coauthored by Kathleen Pryer and Harald Schneider
of Chicago's Field Museum of Natural History and Alan R.
Smith and Ray Cranfill of the UC Berkeley Herbarium, provided
very strong evidence that ferns and horsetails are one another's
closest relatives and the group most closely related to
the seed plants.
"It clarifies one big chunk of the tree," Mishler said.
"We haven't completed the whole tree, but these papers one
at a time have dealt with all aspects of the green part
of the tree of life."
The Green Plant Phylogeny Research Coordination Group,
initially funded for a five-year period by the U.S. Department
of Energy, NSF and the Department of Agriculture, was initiated
by plant biologists as a way to make sense of the reams
of data on plant relationships.
In a series of meetings over the past five years, more
than 200 biologists reached consensus on the most important
plants to target in genetic studies and the best genes to
focus on. Workshops and a Web site clearinghouse for phylogenetic
information helped the community of plant biologists coordinate
research and answer important questions about plant relationships.
"It is important to emphasize that this field used to be
very independent and lab-oriented, where everyone was working
in secrecy within the walls of their lab," he said. "But
as a result of Deep Green, people began to cooperate. They
started sharing data and techniques, and that's where this
progress came from."
Among Deep Green's achievements was completion of a good
draft of the tree of life for green plants. It identified
a cream-colored flower called Amborella as the earliest-diverging
lineage in the flowering plants; concluded that land plants
first emerged onto land from fresh water, not the salty
oceans; and made clear that, at many critical transitions
in evolution, only one lineage of green plant survived.
Such information on plant relationships becomes extremely
important as researchers try to engineer new traits -from
disease resistance to drought tolerance - in crop plants.
With Deep Gene, funded through a Research Coordination
Networks grant from NSF, Mishler hopes to repeat the success
of Deep Green. This time, however, he is bringing in scientists
working on plant genomics to reach consensus on the most
important plants to target for genome sequencing.
The genome sequence of the widely-used research plant Arabidopsis
thaliana is nearly complete, and sequencing of the rice
and corn genomes is underway. Genomic data is publicly available
on some 19 other plants. To make the most of sequencing
efforts, Mishler said, scientists should choose more diverse
plants that cover the range of economically important land
"You have to pick the landmarks. If you want a good representation
of the whole tree of life, you need to pick genomes nicely
spaced on the tree," he said. "Then, for example, once you
understand the genes involved in flower development in one
species, it's not too difficult to probe for the genes involved
in flower development in nearby species."
He notes that the long-term goal of plant genomics is to
identify, isolate and determine the function of genes associated
with various plant traits. This can be facilitated by a
quality tree of life. Using sister group comparisons, for
example, researchers can locate two closely related plants,
one with a particular trait and one without, to help them
reduce the number of genes they need to look at to isolate
those responsible for the trait.
"Ideally, you could narrow the search down to probably
just a few genes from thousands," he said.
Alternatively, ancestor-descendent comparisons allow researchers
to study complex systems of interacting genes, such as those
controlling the angiosperm flower, at a more primitive evolutionary
stage, for example, when they were involved in moss and
One area where this approach has borne fruit is the study
of dessication tolerance, the ability of plants to withstand
drought. If the trait, common in algae, ferns and lichens,
can be transferred to crop plants, they might subsist on
less water or better survive drought.
In a report in last November's Journal of Plant Ecology,
Mishler and U.S. Department of Agriculture researcher Melvin
J. Oliver used sister group comparisons to help unravel
this complex phenotype, which involves more than 80 interacting
"When plants first invaded the land, they were all vegetatively
dessication tolerant - they could dry up completely and
still be rejuvenated. But as plants evolved more complicated
structures, they lost this ability," Mishler said.
"The interesting story is, dessication tolerance re-evolved
at least eight times within flowering plants, and again
when the seed evolved. It appears, from our initial work,
that many of the genes involved in seed dessication tolerance
are descendents of the early genes that were involved in
vegetative dessication tolerance in the first place."
These findings emphasize the value of studying simpler
plants to better understand higher plants, Mishler said.
"We now have real hope that we will be able to understand
something about these economically very important events
in evolution, the evolution of the seed and of the flower,
by looking at mosses and ferns and algae, which are much
simpler study systems," he said.
More insights are sure to come from the interaction between
systematists like Mishler, who chart the evolutionary relationships
among plants, and genomicists identifying the genetic makeup
of green plants.
"Deep Gene is an attempt to meld together the plant phylogenetics
progress we've made rapidly in the last few years with the
rapid progress in plant genomics," Mishler said. "We believe
this will be a truly synergistic process, where genomicists
and phylogeneticists both benefit."