Transcript RG_Talk5PM_Nov6-1

The plant biologists’ goals are to…
> communicate desired outcomes of visualization
> generate tools that unite data sets
> address the genoytpe<>phenotype conundrum - G2P
enable new insights into processes that are rendered
in visually intuitively accessible form
Investing Images of Biological
Processes with Added Meaning
• Events occurring in time and space, in three
dimensions.
• Relationships between different levels of
biological organization/data types.
• Transforming images by populating them with
quantitative, dynamic, data.
Questions Plant Biologists Ask
• 1. What chains of changes occur in a plant over
time and space in response to alterations in the
environment or “signals”? Drought, Cold, Light
• 2. What internally “programmed” chains of
changes in a plant occur over time and space ?
Flowering, getting old, beginning photosynthesis.
• 3. How are these events regulated and what
interactions occur between 1 and 2? That’s why
we need more sophisticated, and quantifiable,
images.
Overarching Goal and Difficulty
• Goal: To find dependable ways to unite the data
available at various scales into coherent pictures
for particular plants in particular contexts.
Generalizations/hypotheses may follow.
• Difficulty: There are no complete data sets, and
the mechanistic bases for the data that are
available are not fully understood. Intelligent
guesses are the norm! We want to create a
context in which those guesses can become more
accurate.
The Current Terminology
• The term “plant biology” is used to avoid
categorization into scales and modes of
measuring the data.
• A unifying goal of the plant biology community is
to reveal and make clear the underlying cellular
events (“mechanisms”) that give rise to a
particular outcome (a “phenotype”).
• This is ultimately proved by establishing a causal
connection between “genotype” and
“phenotype”. Not easy.
Genes and Genomes
• Genes contain information. They are encoded in
the molecule DNA (deoxyribonucleic acid).
• Genomes contain all the DNA in an organism.
Different information is accessed under different
circumstances, “readout”.
• Not all of that DNA is in the form of genes.
• The ancestral relationships of organisms are
manifest in the relatedness of their genomes.
Cells and Genomes
• All organisms are composed entirely of
cells, or the products of cellular activity.
• With some exceptions, all cells contain a
complete copy of that organism’s genome.
Diagram of a Plant Cell
Plant cells have internal compartments or organelles.
Cytoplasm- site
of translation
Each
organelle
serves a
distinct
function and
is
surrounded
by a barrier
or
membrane
Nucleus –
majority of
the plant
genome
Chloroplast - site
of
photosynthesis.
(Tom will show
data about
photosynthesis.)
The plant genome is divided among the nucleus, the chloroplast and the
mitochondrion. Readouts from the three organelles are coordinated.
Genes and Genomes 2
No two genomes are identical, even within a
species. (Genetic variation)
• Each genome is also a historical document,
illustrating past evolutionary processes. We
cannot completely read the document, however.
Genotype, Phenotype?
• A fundamental distinction has to be made
between all the genes possessed by an
organism (Mendel’s “heritable units”), its
“genotype” AND
• The characteristics that it exhibits, or its
appearance, its “phenotype”. Phenotype is the
manifestation of a particular readout of the
genome.
“Phenotype”
“Phenotype” means appearance. It originally meant
“appearance” as it appears to the human eye of the
observer e.g. the surface of Mendel’s pea seeds or
fruit fly eye color.
Nowadays it can also mean any manifestation of
genome readout, many of which are observed with
the help of contemporary, high-throughput
technology, as “read” by computational tools.
Plants Have A Special Problem
Versus
I’m out of here!
Running away is not an option
External and Internal Signals
Sorry, Tom, you’ll have to pretend it’s a corn plant!
Signal is
perceived at
the cellular
level .
Genome
responds
Signals
Light
One kind
of readout
Internal Developmental Signals
Buchanan, Gruissem, Jones, 2000, Biochem,. Mol. Biol. Of Plants (BGJ)
Scales Over Which Data Are Read
• 1. Whole fields of plants, or whole plant behavior (Crop
Science). (Steve W., Jeff White, Jim Jones)
• 2. Individual tissues, e.g. roots or leaves, (Plant Physiology).
(Steve W., Siobhan Brady).
• 3. Cell types, e.g. mesophyll cells or bundle sheath cells. (Tom
B.)
• 4 The genome, transcriptome, proteome, metabolome.
Language of “omics”). (Nick P., Tom B., RG, Siobhan B., All of
us, ultimately)
• Note: Individual genes come in different “flavors”, called
alleles, giving rise to slightly different mRNAs, and slightly
different proteins, but still with the same function
The Language of “
”
“omics” alludes to the study of the totality of
information associated with defined levels of biological
organization. (This classification is way over-simplified).
“Genomics” – genes, structure/”landscape” of the
genome.. A constant for any given organism.
“Transcriptomics”- transcripts. Variable
“Proteomics” – proteins. Variable
“Metabolomics”- metabolites, Variable
“Phenomics”- phenotypes. Variable
Different Views of Genomes
• 1. Some people want to know about the history of
genomes, how they came to have a particular
structure, and, from that, how different organisms
are related. (Eric L.)
• 2. Others want to know how the information in the
genome (”genotype”) is variously read so as to
produce a specific organism that grows, reproduces,
and responds to internal and external signals
(“phenotypic plasticity”). (RG, TB, SW, NP, BU).
Genome Function
Would that it were that simple!
The readout cannot really be described as simply proceeding from the first level of
transcription because there are many other points of control.
Complete
information
Readout from
the genome
Different genes are
under different
conditions.
Different proteins are
present under
different conditions
Translation from the
language of
DNA/RNA into
protein
Proteins: the
workhorses
of the cell
Harder Solutions
• The era of “blame it all on one gene/allele” is
coming to an end, although it still holds in some
cases.
• We now know that any particular phenotype may
result as the outcome of a network of interacting
genes. (Actually, the readout from those genes.)
• That interaction may be manifested between
genes or protein molecules, and always at a
specific location and under a specific condition
within the plant cell.
Different Levels of Genome Regulation
The amount of
mRNA
available for
translation can
be varied
The
amount of
protein
can be
varied
mRNA
The protein
can be
chemically
modified in
different
tissues.
Functions of Proteins
• The proteins that are produced perform functions such as
– Catalyzing chemical reactions that would not otherwise go on in
the cell, producing “metabolites”.
– Acting as structural components of the specialized membranes
in the cell.
– Causing the activation of particular genes by binding to their
promoter region, “transcription factors”.
– Interacting with each other in a “signal cascade”.
• For each of these categories, “There’s a database for that”,
but they are not yet fully integrated, nor fully visualized.
Nick will show us a tool that integrates databases.
Transcription Factors (TFs)
DNA binding
proteins often
interact with
each other.
/ TFs
The TFs bind to
sequences that
are specific for the
gene
The expression of
groups of genes is
often
coordinated.
Much active work
in analyzing
patterns of gene
expression in
response to given
stimuli (Nick, RG,
and many
others).
“Promoter region”
Gene is transcribed to mRNA
after proteins have bound to
adjacent DNA, “promoter”.
External Signals
Signal acts
on the
genome
though TFs
A “Signal Cascade”
Transmission of a
signal, from, say,
the periphery of a
cell to the nucleus,
where a particular
gene (s) is then
activated.
This mechanism
involves the
physical
interaction of
proteins with
each other.
Databases of
known
protein-protein
interactions are
available. (Nick’s
work on “The
Interactome”)
Simplified carotenoid biosynthetic
pathway in plants
Example of a
metabolic pathway
Caps and
bold:
enzymes
Tom and his
colleagues studied the
effect of genetic
variation in maize on
the levels of the
nutrient compounds
in red.
The variation is due to
small differences in the
genes that encode these
enzymes.
C. E.
Published by AAAS
Metabolic
Pathways
Many metabolic
pathways involve
physical
interactions
among enzymes.
Stress hormone
Harjes et al., Science 319, 330 -333 (2008)
Metabolic Pathways: “MapMan”
• Bjoern will show us a tool for the depiction of
metabolic pathways in plants, and for pasting
gene expression values on to those pathways.
In MapMan, genes are “binned” into
pathways.
What is Missing
• A way to integrate these phenomena so that the
user can visually follow known events from,
say, the appearance of the signal all the way to
the phenotype passing through the
corresponding chain of events.
A Crossroads for Plant Biology..
There are models and then
there are models
“…….models bring together two of biologists’ favorite tools:
abstraction and a love of drawing. Often they indulge in both
of these passions by sketching down simplistic models, which
may include five arrows, a factor X, and two question marks.
What has worked for decades to illustrate interactions
among a few genes, has clearly reached its limits in the postgenomic era where research is no longer data-limited.”
Sensing
Signaling
Salinity
Drought
Temperature
Ionic
Responses
Stress-specific
Homeostatic
Adjustment
Osmotic
Temperature
*
Growth Control/
Development
Injury
Status
***
Cell
Death
Detoxification**
Damage Control
& Repair
T
O
L
E
R
A
N
C
E
or
A
V
O
I
D
A
N
C
E
SOSs, DREBs, CBFs, NACs
* protein unfolding, membrane leakage, water/ion imbalance;
** distinct pathway, not specific for a particular stress;
*** overlapping, pathways
acc. Zhu, 2001, Bohnert & Bressan, 2001; Grene & Bohnert, 2009
[important stress response transcription factors]
are affected by and themselves affect
ROS- and redox-dependent
pathways
The Study of Genome Function is
• 1. We only know the full sequence of a few genomes.
• 2. Even for those in which the sequence is known, we do
not have a full picture of how the genome is “read”.
• 3. Valid generalizations about genome structure and
genome readout can be made, but, mostly, we cannot
identify all the “working parts” for any organism.
• 4. Some biologists draw “models”, which usually means a
static pictorial summation of what we know about a
given phenomenon. No ODEs or stats involved, unless
you are Steve Welch and the like.
MYB96 mediates ABA signals via RD22 in
regulating drought stress response
An early response to drought is the synthesis of ABA
The TF activates
this gene
When
drought
occurs,
ABA is
made,
causing
the TF
gene to be
expressed
Phenotype
Expression of this
TF
Phenotype
Seo, P. J., et al. Plant Physiol. 2009;151:275-289
ABA is a plant hormone, produced under stress
A “Model” of ABA-Dependent and
Independent Stress Responses in Arabidopsis
Inhibition by these
proteins
TFs binding to
promoter of
this genes
Everything beyond
expression of this gene is a
black box.
A “Model” for Temperature Responses
in Arabidopsis
Grey ovals contain TFs that respond to more then one stress
TFs
(ABA, GA, auxin, ethylene are plant hormones).
The rectangles contain generalized allusions to phenotypes The
arrow leading to each rectangle is a type of hand-waving.. Time
and space are hard to decipher here. Can we improve on this?
Wish List: Layering
• A system of layers, where the user can mine a given set of
data/”layer” (repositories of gene expression, metabolite
data, protein-protein interactions), and view those data in a
cellular context in 3D. A suite of genes/proteins/metabolites
could be visualized dynamically, showing changes over time.
Layers could be fused at will, within a temporal or spatial
context.
• The user could upload his/her own data.
• All data would be linked to other available information, for
example, orthologs in other species, published diagrams of
signaling pathways, what is known about regulation of the
genes, proteins or pathways of interest…..
Each Sub-Discipline has its Own
Manhattan, NY
Rallying Cry
Unite the
languages, enable
a view that yields
quantitative data
from fields of
chilly plants to
cells where
signaling
cascades operate
to activate
defense genes, to
altered
metabolism to
chilling-tolerant
plants.
The Flyovereverybody else
“Center of the
Universe”, e.g. Fields,
Tissues, Cells,
Genomes
Bjoern Presents His Data
A “Model” of the Regulation of ABA
Signaling by AREB1 (a TF)
Stress Response
Fujita, Y., et al. Plant Cell 2005;17:3470-3488
Genome Readout 3
• Production of mRNA and its regulation
• Production of protein and its regulation.
• Production of functionally active protein in a
specific subcellular location.
• Mechanisms for providing other factors
needed for function (e.g. substrates for
enzymes).
• Effect of the functioning of proteins on
cell/tissue/ plant function.
Making the G2P Connection
Connecting a particular allele to a particular
phenotype. How is gene expression
regulated?
Connecting a substring within a given
chromosome to a given phenotype. Stats.
Connecting a gene network to a given
phenotype. How is the network regulated?
Connecting a metabolic pathway to a given
phenotype. How is the pathway regulated?