Phenome Network demonstration

1
Phenome Network - demonstration

• URL link
• http//phn.huji.ac.il/cgi-bin/RTQ/index.pl

A system for the analysis of quantitative
phenotypes on both plants and animals for
breeding and research
This presentation will demonstrate the use of
Phenome Networks using the tomato pennellii
introgression lines (ILs) population that was
developed in Prof. Dani Zamir lab.
2
(No Transcript)
3
First, point on the Organism menu to select an
organism (now you dont need to do it because
there is only Tomato, but later versions will
include more organisms).
4
Then point on the Populations menu to select a
population (now you dont need to do it because
there is only M82-pennellii ILs, but later
versions will include more).
Now that you selected a population from a
particular organism, you are ready to start
exploring the phenotypic data that is available
for this population...
5
You can start by viewing the general information
about the population (click on population
info), browsing the experiments (click on
Experiments), browsing traits (click on
Traits) and browsing the genotypes (click on
genotypes).
6
For example, if we click on Experiments we get
this table that lists all experiments that are
available for the ILs.
7
Here is an experiment card example for Akko
2006.snif I experiment.
8
The same is true for traits. Here is a list of
all available traits
9
This is a trait card for b_alanine.
10
But lets start the analysis from the beginning.
Suppose we are interested to analyze a particular
trait in a particular experiment.
11
This first menu shows all available experiments
12
After you selected experiment (for example akko
2004.ILs) you get a list of all traits measured
in this experiment. Now lets select a trait that
we want to analyze. For example Total Yield
(TY).
This will take us to a tour to explore TY in
Akko 2004 exepriment ..
13
As a start, we get this frequency distribution of
all genotypes measured for TY in this
experiment. On the right there is small table
that shows basic statistics on the trait.
As is true for every figure, you can download it
to your computer by clicking on the image with
the right button of the mouse and select save
picture as.
14
This is the result of displaying both homozygous
and heterozygous
You can choose to perform the distribution based
on homozygous genotypes only, heterozygous
genotypes only or display both homozygous and
heterozygous together.
15
Scrolling down youll get the possibility to see
this distribution in other experiment in which
TY was measured or for other trait that was
measured in Akko 2004 experiment.
Another option is to correlate homozygous and
heterozygous genotypes for this trait
16
Here is the correlation between homozygous and
heterozygous ILs. Each data point represents an
IL where its x axis is the (mean) homozygous line
(percent of control) and the y axis is the
heterozygous line.
17
Here is the comparison between homozygous and
heterozygous ILs. Each data point represents an
IL or ILH where the y axis is the genotypes
average. The fact that the homozygous name is
colored in red and heterozygous is not colored
means that the difference is statistically
significant (otherwise both of them would be red).
18
This gives a bar plot in which each bar is a
genotype, and its Y axis is its average. Green
bars are homozygous genotypes and red are
heterozygous. You can point on the bar to see the
genotypes identity.
19
Here is the result
20
Genotypes order by chromosome
21
This is a table of general statistics about each
genotype for TY. it has mean column, rep (
of replications), std_dev (standard deviation)
and mean_percent (percent of control). You can
click on the genotype name to go to genotype
card page.
22
Which gives you this table
Columns interpretations IL mean average of
the homozygous IL, ILWT ttest ttest
comparison between IL and M82, ILH mean mean
of the heterozygous IL (ILH), ILHWT ttest
ttest comparison between ILH and M82, ILILH
ttest ttest comparison between IL and ILH.
MOI_qual mode of inheritance for this QTL R
recessive, A additive, D dominant, ODO
overdominant, increasing QTL, - decreasing
QTL
Now click on the genetic map
23
This is the genetic map of the ILs for TY
(Total Yield). Each vertical thick blue line
represents a linkage group (chromosome), and the
vertical lines on its left side are the ILs.
There are 2 adjacent vertical lines to each IL
the left is the homozygous and the right is the
heterozygous. They are color coded according to
red genotype is significantly higher than M82
in TY, blue genotypes is significantly lower
than M82, black genotype is not significant
than M82, grey genotype has not been measured
for this trait.
24
Here is the comparison for the M82, IL1-3,
ILH1-3. again, each data point represents a
plant, and its clickable, so you can see trait
profile for this plant.
25
Or you can click on a chromosome to compare all
genotypes associated with it.
26
Now genotypes are colored quantitatively
according to this bar, such that intense red is
high, white is like M82 and blue is low.
27
This for example is the result of selecting the
first 4 genotypes
Figure explanation This is a F test Tukey
comparison among all selected genotypes. Each
diamond represents a genotype (that is written on
the bottom), such that the line on the middle of
the diamond is the genotypes mean, and the
height of the diamond is its confidence interval.
Each data point represents a single plant
(replication) that is part of a particular
genotype and its y axis is its total yield
value. M82 is highlighted in red, and all the
rest of genotypes are compared to it, such that a
red genotype means its not significant and a
black means its significant. Here IL1-1-2-I is
the only one that is significantly lower than M82
28
This is a traits profile for the selected plant.
Each blue vertical line represents the variation
(CV) of a particular trait (that is written
below), and the red circle represent the value of
the selected plant for this trait as percent of
control
29
Lets continue with the next option. Here you can
select a trait (without the need to select
experiment), so you can view the consistency of
this trait among all experiments. Lets select
morphology
30
The list of all morphology traits appears. Then
we can select (again) TY (Total Yield).
31
And this is what we get
Figure explanation each blue square on the
diagonal indicates an experiment. Every non-grey
square represents the correlation between the 2
corresponding experiments red (or pink) square
indicates a significant correlation (more red
more significant) while white indicates non
significant correlation. Grey square means there
are no shared genotypes between the corresponding
experiments. Point on each square to see the
correlation parameters (R coefficient, N
sample size, P p value), while click on the
square give you the correlation plot.
32
Here is the correlation between the 2
corresponding experiments (in this case akko
2000 and akko 2006).
Figure explanation each data point represents a
genotype, while its X axis is its (mean) TY in
the first experiment, and the Y axis is its mean
in the second experiment. Point on each data
point to see the genotype.
33
And here you can click on each data point to see
the traits profile of this individual (plant).
34
Getting back to here, you can select an IL to see
its performance of TY in all experiments.
35
This IL1-4 performance in all experiments for
TY. Purple bar is the M82, red bar is
heterozygous IL (ILH) and green bar is homozygous
IL. Experiment name is written on the bottom of
each panel.
36
Lets continue with the next option. Here you
select an experiment.
37
This is a heatmap of all phenotyes and genotypes
in the experiment. The Y axis list all genotypes
and the X axis all phenotypes. A red square
indicates that this genotype has high phenotypic
value, blue means its low phenotypic value and
white means average.
38
Another possibility is to view QTL numbers for
all traits measured in this experiment by
clicking here
39
this figure indicates the number of QTLs for each
trait (vertical bar), such that bars higher than
the 0 line indicates increasing QTL and lower
than 0 line indicates decreasing QTLs. They are
also classified into mode of inheritance
categories that are indicated by the different
hues of the bars (see legend on the bottom left
part. Point on each bar shows the trait name and
clicking on it will bring you to the genetic map
figure.
40
Going back to the heatmap figure you can select
to view correlations among all traits in this
experiment.
41
This gives us another heatmap, although that
here, both the X and Y axes represent traits, and
each little square indicates the correlation
between the corresponding traits. Red means
positive correlation, blue negative and white
no correlation. Both axes are clustered according
to similarities among traits.
42
Well continue to the next option. Here you can
correlate among all traits in the system. For
this press Go.
43
This page shows all traits in the system,
organized according to source (researcher),
category and type. You can select one or more
traits, and then press on Go (on the right
bottom panel of the page) to show all
correlations of the selected traits to all other
traits in the system.
44
We get this table, in which each row represents a
correlation between TY of a particular
experiment and BX of a particular (possibly
different) experiment. For example, the first row
shows the correlation between BM (var1 column) in
experiment akko 1993.ILs (experiment_name1
column) and TY (var2) in the same experiment
(experiment_name2 column). The columns R, N
and P indicates the correlation parameters. A
click on each trait or experiment name brings you
to the card of this trait/experiment. You can
also click on plot to see the correlation
figure.
45
Suppose we want to see correlations (in all
different experiments) of only TY and BX
(brix). For this we first need to select these 2
traits. But immediately on Go will give the
correlation of TY and BX to all other traits
and not between each other. So for this we need
to change the parameters in the form of the
right. Just go to how to treat selected traits
and choose the correlate selected traits with
each other option.
46
You get this table such that each row shows a
correlation between TY and BX in 2
experiments.
47
Here is the correlation figure. Each data point
is a genotype where its X axis is its mean in
BX and its Y axis is the mean of TY (both
units are percent of control). Green points are
homozygous and red are heterozygous. Point on
each data point to highlight the genotype.
48
Here are all correlations between TY and BX
in all experiments. point on each panel to see
the correlation parameters, and click on it to
focus on that panel.
49
Well continue to the last option, which is
Field block. This options is accessible from
the experiments page. Click here to continue.
50
This takes you to the field block menu. Now we
need to select an experiment (for example akko
2006), and then a block number (block_1), and
then a trait (TY).
51
This is the actual block in the field in Akko
2006 experiment. Each colored square represents a
plant, and its color indicates its phenotypic
value (according to the color bar on the right).
Such that red square are high yielded plants and
blue are low yielded plants, and white are
average. You can point on each square to see the
plant details and its phenotypic value, or you
can click on the F2 family on the top (for
example IL8-3 on the right), to compare between
the IL, ILH and M82 in this family.
52
Here is the comparison for TY between the 3
genotypes of IL8-3 F2 family. Again, each data
point represents a plant that can be clicked on
to see it phenotypes profile.
53

• all analyses that are demonstrated in this
  presentation are performed in real time.
• this means that every set of data on
  introgression lines (in any organism) can be
  uploaded to the system. Once its done all
  analysis demonstrated here will be available on
  the new data set.
• Later version of Phenome networks will include
  the ability to analyze also population of
  different genetic structures like recombinant
  inbred lines and F2 as well as breeding programs
  data.
• any question/comment can be sent to
• semel_at_agri.huji.ac.il