| Preliminary Single Environment Analysis |
| See Also |
The QTL analysis in GenStat currently requires trait means rather than individual
plot (or unit) values. This menu provides facilities for a preliminary single environment
analysis using mixed models to predict trait means for that environment. These predicted
means can be saved, either alone for a single trial, or in combination (stacked) with means from
other trials (environments) in the case of multi-environment trials (METs).
The menu provides minimal facilities for model building and investigation - this should
be done beforehand (use the appropriate mixed model (REML) menu).
The preliminary single environment analysis runs two REML analyses. In the first
analysis the genotype factor is fitted as a random term - in order to estimate variance parameters
and heritability - and in the second analysis it is fitted
as a fixed term - in order to obtain unshrunken predicted means. By default, the second run
uses variance parameters estimated in the first run, but this can be changed using the Options menu.
Data
Specifies the response variate for analysis.
Environment
Provides a drop down list of the environments available for use in the analysis. In the case of a
single environment trial there will be one item in the list called
single environment. For a multi-environment trial,
if an environment factor has been
specified in the QTL workbook then the levels or labels from this factor will be
displayed in the list. Alternatively, an environment factor can be selected by
clicking the select factor button. When an environment is
selected the menu will create subsets of data to use for the analysis.
Genotypes
These options give information on structure within the Genotypes factor. In the simplest case,
all of the genotypes in the trial are to be included in the QTL analysis, and in this case
the default options are adequate: select the
all genotypes in QTL analysis option and specify the name of the factor in the
Genotypes input field.
Alternatively, if there are some genotypes present in the trial that
are not to be used in the QTL analysis, such as check lines or parents, select the
extra genotypes present option. When this option is selected the
genotypes need to be supplied using two factors. If you have a single factor containing all the genotypes
present in the trial, these factors can be generated by clicking on the
create genotype factors, or alternatively you can supply these factors directly.
The first factor, supplied using the
genotypes input field, has levels/labels corresponding to the genotypes to be used
in the QTL analysis, with missing values used to represent the extra lines.
The second factor, supplied using the extra lines field, has levels/labels
corresponding to the extra lines plus an additional level/label (e.g. level 0 and/or label 'Test line')
used to identify the set of genotypes taken forward for QTL analysis. For example, the following table shows how
the levels
of two factors could be specified for 22 genotypes where levels 21 and 22 represent the 2 parents:
| Test line factor | Check line factor |
| 1 | 0 |
| 2 | 0 |
| 3 | 0 |
| ... | ... |
| 20 | 0 |
| * | 21 |
| * | 22 |
Note that it is usually desirable to analyse the whole trial then extract predicted means to take forward to
QTL analysis as opposed to analysis of a subset of the trial.
Design
Selects the type of design used in the trial. When a design is selected the menu
will change to display appropriate input fields. Click on the design for a description
of the associated input fields.
Allows you to fit a general mixed model.
-
Additional Fixed Model
In this context, the additional fixed model specifies treatment factors and/or covariates for which the predicted genotype
means are to be adjusted. The predicted genotype means taken forward will be calculated as an average value
across factor levels, or at the mean of a covariate. This model is written using a formula, which can combine main
effects and interactions of factors and also covariates. The Genotype factor should not
be specified here, as it is added to the model automatically.
Additional Random Model
In this context, the additional random model is used to describe the physical structure (blocking) of the trial.
The model is described using a formula, which can
combine main effects and interactions of factors and also covariates. The Genotype factor should not
be specified here, as it is added to the model automatically.
-
Blocks
The block structure is described using a formula.
-
Blocks
The block structure is described using a formula.
-
Row factor
A factor specifying the rows of the regular grid.
Linear trend across rows
Specifies whether you want to fit a linear trend across rows.
Row-model
List the available correlation models. Select the correlation model that you want to apply across the rows of the grid.
Random row term
Specifies whether you want to fit a random row effect in addition to the row-model.
Column factor
A factor specifying the columns of the regular grid.
Linear trend across columns
Specifies whether you want to fit a linear trend across columns.
Column-model
List the available correlation models. Select the correlation model that you want to apply across the columns of the grid.
Random column term
Specifies whether you want to fit a random column effect in addition to the column-model.
Additional Fixed Model
In this context, the additional fixed model specifies treatment factors and/or covariates for which the predicted genotype
means are to be adjusted. The predicted genotype means taken forward will be calculated as an average value
across factor levels, or at the mean of a covariate. This model is written using a formula, which can combine main
effects and interactions of factors and also covariates. The Genotype factor should not
be specified here, as it is added to the model automatically.
Additional Random Model
In this context, the additional random model is used to describe any additional physical structure (blocking) of the trial.
The model is described using a formula, which can
combine main effects and interactions of factors and also covariates. The Genotype factor should not
be specified here, as it is added to the model automatically.
Available Data
This lists data structures appropriate to the current input field. Double-click on a name to copy it
to the current input field; alternatively, you can type the name directly into the input field.
When data has been stored in a QTL workbook then by default the available data
will display only the data structures present within that workbook. You can right-click on the available data
list to switch between displaying data only within the QTL workbook and all data within GenStat. If the QTL
workbook has not been used, then all current data structures within GenStat will be displayed.
Operators
This provides a quick way of entering operators in the fixed and random
model formulas. Double-click on the required symbol to copy it to the current
input field. You can also type in operators directly. See
model formula for a description of each.
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 shortcut menu where you can choose to set
the menu using the currently stored defaults or the GenStat default settings. |
| Further output | Opens a dialog for specifying further output from the analysis and
displaying residual plots. |
| Save Test Line Means | Opens a dialog where you can save genotype means for a QTL analysis. |
See Also
Show All