|
Introduction
ClassMarker
operates on expression data measured on a number of genes in samples of
different classes of cells. A class may be a particular tissue (e.g.
brain, muscle, tumor), a particular disease status (e.g. normal,
diseased, a particular stage of a disease), a particular disease
(e.g. various types of cancer), etc. The membership of each sample
in its class must be known with as high a degree of confidence as possible,
for instance on the basis of histology, the known evolution of the disease,
or other means of diagnosis.
The data supplied to ClassMarker thus form a matrix
where each row corresponds to a gene and each column to a sample. An element
in the matrix contains the expression level of the row’s gene measured
in the column’s sample. Each of the columns is furthermore classified
into one of the sample classes. In
order to be effective, the method used in ClassMarker needs data from
as many samples in each class as possible. It is worth noting that the
assignment of samples to classes is easily specified with the rich graphical
interface available in ClassMarker (learn more on class
creation and experiment reordering).
Provided enough samples of known class are supplied,
unclassified samples (samples of unknown class) may be supplied
as well, in form of additional columns. ClassMarker will attempt to classify
those samples into one of the known classes.
When the data are read in, various filters
can be specified (e.g. min/max expression level, minimal fold change,
logarithmic transformation). This may modify the expression values
taken into account, or eliminate genes from the analysis. The impact of
the filters is conveniently monitored by the graphical interface.
Once the filters are specified, the proper analysis
can start. Two types of analysis are available:
- identification of marker genes and assessment of their quality by
cross-validation
- identification of marker genes and assessment of their quality by
train-and-test evaluation.
In both cases, marker genes are
identified on the basis of a subset of the classified genes and used to
classify the rest of the genes – the quality of the markers is then
assessed by the success in classifying each sample into its proper class.
For the latter stage of classification, two classification techniques
are available:
- a proprietary voting method
- a k-Nearest-Neighbors classification.
While the choice of the classification
technique depends on the data, the first is usually the best. When it
is used, it is possible to take into account couples of classes in identification
of the markers and the subsequent classification.
ClassMarker offers a unique graphical interface
that allows for a deep analysis of the data. Individual samples can be
excluded from the analysis, classes can be merged and split, etc., with
unparalleled ease. This enables the scientist to test many hypotheses
and identify promising target genes in record time.
How
to Proceed
1) Load your data with one of the following choices:
2) Create your classes
- Specify filter parameters
if necessary
- Reorder your data if necessary.
- Create the different classes.
3) Start an analysis
4) Analyze your result
- Switch to the freshly created Result
Tab.
- Use the different views in the Result Tab to analyse the result.
5) Add other analyses
- Add other analyses if necessary, by repeating the sequence from step
3. Various scenarii are easily obtained by
Note that modifying the Master Tab in order to run an analysis with
a different scenario does not invalidate the results obtained with
previous scenarii, since all the necessary information is stored with
the result in each Result Tab. This allows for a very easy comparison
of results obtained with various scenarii by simple switching from
one Result Tab to another.
6) Save your solution
- Use the Save button of the toolbar
or the Save command in the file
menu to save your project as an AMP file. An AMP file contains all
the details of a ClassMarker run, when it is subsequently reloaded into
ArrayMiner, ClassMarker is set into the state in which the file was
saved.
|
