This menu can be used to estimate effects from the within slide differences between targets. This
information is contained in the log-ratios. Usually, these log-ratios will be normalized
using the Normalize Two Channel Microarray Data
menu.
Once the corrected log ratios, Mij (where i indexes the slides and j the probes), have been obtained, the
analysis of the differences between targets is performed using this menu.
This uses analysis of variance (ANOVA) with a pooled error across treatments/targets for each probe.
The normalization of each slide effectively removes the block effects, so the
Mij values now reflect the differences between treatments on each slide
and there is one measurement per slide rather than two. Thus the design matrix, X,
for the ANOVA is not standard. If the t target effects are given by
T = [T1, T2, … Tt],
then the model in matrix notation is:
The constant term μ estimates the dye swap effect for the probe. The usual restriction on T is
required to obtain estimates, as one parameter is redundant. If we impose the restriction, then
we have the usual least squares solution:
where X1 = [1...1]' and
where R is the (t+1) vector [0, 1…] that applies the restriction above.
Available Data
This lists data structures appropriate for the edit box which
currently has focus. You can double-click a name to enter it in the
edit box.
Data Format
The data can be supplied in either of the following formats:
- Single Variate for Log-ratios with Slide Factor - All the log-ratios are stacked
into a single variate, with factors that index the slide and probe/gene
- Pointer to Log-ratio Variates for each Slide - Each slide has its data in
a variate, and a pointer which points to this set of variates is provided. The Slides
factor is not required, but if supplied it should just have one entry for each slide in the order of
the variates in the pointer. The Probes/Genes factor is that for a single slide, and
all slides must have a common layout.
The spreadsheet stack and
unstack menus can be used to reorganise the data
between these two formats.
Log-ratios
A variate containing the log-ratios to be analysed, or a pointer to a set of variates
for each slide. The variates must all contain the data in the same order.
Probes/Genes
The factor that identifies the probes or genes. If the data are in pointer format, this
factor should contain just the information for a single slide, and all the slides log-ratios are expected
in this common order. If the data are in variate (stacked) format, this factor indexes the probes
in the log-ratio variate.
Slides
The factor that identifies the slides. If the data are in pointer format, then this
factor should contain just one entry per slide. If the data are in variate (stacked) format,
then this factor indexes the slides in the log-ratio variate.
Red Treatments
A factor containing the target assigned to the red/Cy5 dye. This is assumed to be the channel on the top
of the log-ratios. This factor must have the same number of values as levels in the Slides factor.
Note: Cy5 is the technical name for the red fluorescent dye.
Green Treatments
A factor containing the target assigned to the green/Cy3 dye. This is assumed to be the channel on the bottom
of the log-ratios. This factor must have the same number of values as levels in the Slides factor.
Note: Cy3 is the technical name for the green fluorescent dye.
Slide Order Validation
A factor the same length as the Red Treatments factor which indexes the slides.
This must have the same levels/labels as the Slides factor, and is used to
verify that the treatments are in the same order as the data specified in the log-ratios.
Supplying this factor is optional, but highly recommended to validate whether the data and
treatments match as expected. If the labels of the slides and check factor match, but are in a
different order, the treatment factors will be sorted into the correct order with a warning.
Contrasts
The optional name of matrix specifying contrasts to be estimated. The contrasts matrix
must contain a column for each treatment/target and the sum of the rows should equal zero.
To create a new contrasts matrix, use the Contrasts button.
Action Buttons
| Run | Run the analysis. |
| Cancel | Close the menu without further changes. |
| Options | Opens a dialog where additional options and settings can be
specified for the analysis. |
| Defaults | Set the menu settings back to the default settings.
Clicking the right mouse on this button produces a pop-up menu where you can choose to set
the menu using the currently stored defaults or the GenStat default settings. |
| Store | Opens a dialog to specify names of structures to store the results from the analysis.
The names to save the structures should be supplied before running the analysis. |
| Contrasts |
Creates a contrasts matrix. If either the Contrasts or Red Treatments fields
are blank then you will be prompted to supply the names for these. Once the names of the contrast matrix and
red treatments factor have been supplied you will be prompted for the number of
contrasts, and then a spreadsheet will be created where you can enter the names and values for the
contrasts.
|
Example
The following shows the estimated effects from a mouse knock out experiment.
The corrected log-ratios from the normalize two channel microarray data
menu have been saved in cLogRatio. A spreadsheet containing the structure of the experiment
has been set up:
The trial contains 16 slides, with 8 knock mice and 8 normal (Control) mice compared
with a standard reference. The menu to estimate the effects is set up as follows:
Note that the column SlideName is set up to check that the order of the labels
in the factor Slide is the same as the order of the slides in this spreadsheet.
A contrast matrix has been set up which provides the comparison between the knock out
and control mice. To create this spreadsheet you can use the Contrasts
button, and specify a single contrast. However as there are no dye swaps in this experiment,
the Red Treatments factor does not contain all the treatment labels, so that you
will need to add a column to the created matrix. The row name and label have also been
edited to change these from the default values, and the 1 and -1 have been entered
for the Knock Out and Normal groups giving the contrast Knockout - Control.
The options for this menu are set with the Options dialog shown below. Note that
as there are no dye swaps the Estimate Dye Bias from dye swaps
option can not be used.
The Store dialog is used to save all the results to a spreadsheet:
This results in the following spreadsheet being produced:
See Also