classifyLDAsc: Function to do discrimination analysis, by the search and...
In maigesPack: Functions to handle cDNA microarray data, including several methods of data analysis

Description Usage Arguments Details Value Author(s) References See Also Examples

Function to search by groups of few genes, also called cliques, that can discriminate (or classify) between two distinct biological sample types, using the Fisher's linear discriminant analysis. This function uses the search and choose method.

1
2
3

classifyLDAsc(obj=NULL, sLabelID="Classification", func="wilcox.test",
              facToClass=NULL, gNameID="GeneName", geneGrp=1, path=NULL,
              nGenes=3, cliques=100, sortBy="cv")

`obj`	object of class `maiges` to search the classifiers.
`sLabelID`	character string with the identification of the sample label to be used.
`func`	string specifying the function to be used to search by the initial one-dimensional classifiers, like 'wilcox.test' or 't.test'.
`facToClass`	named list with 2 character vectors specifying the samples to be compared. If NULL (default) the first 2 types of sLabelID are used.
`gNameID`	character string with the identification of gene label ID.
`geneGrp`	character or integer specifying the gene group to be tested (`colnames` of `GeneGrps` slot). If both `geneGrp` and `path` are NULL all genes are used. Defaults to 1 (first group).
`path`	character or integer specifying the gene network to be tested (`names` of `Paths` slot). If both `geneGrp` and `path` are NULL all genes are used. Defaults to NULL.
`nGenes`	integer specifying the number of genes in the clique, or classifier.
`cliques`	integer specifying the number of cliques or classifiers to be generated.
`sortBy`	character string with the field to be sorted. May be 'cv' (default) or 'svd'.

This function implements the method known as Search and choose proposed by Cristo (2003). If you want to use an exhaustive search use the function classifyLDA.

This method uses the function lda from package MASS to search by classifiers using Fisher's linear discriminant analysis. It is possible to search classifiers by Support Vector Machines and k-nearest neighbour classifiers using the functions classifySVMsc and classifyKNNsc, respectively.

The result of this function is an object of class maigesClass.

Elier B. Cristo, adapted by Gustavo H. Esteves <gesteves@vision.ime.usp.br>

Cristo, E.B. Metodos Estatisticos na Analise de Experimentos de Microarray. Masther's thesis, Instituto de Matematica e Estatistica - Universidade de Sao Paulo, 2003 (in portuguese).

lda, classifyLDA, classifySVMsc and classifyKNNsc.

## Loading the dataset
data(gastro)

## Doing LDA classifier with 2 genes for the 6th gene group comparing
## the 2 categories from 'Type' sample label.
gastro.class = classifyLDAsc(gastro.summ, sLabelID="Type",
  gNameID="GeneName", nGenes=2, geneGrp=1, cliques=10)
gastro.class

## To do classifier with 3 genes for the 6th gene group comparing
## normal vs adenocarcinomas from 'Tissue' sample label
gastro.class = classifyLDAsc(gastro.summ, sLabelID="Tissue",
  gNameID="GeneName", nGenes=3, geneGrp=1, cliques=10,
  facToClass=list(Norm=c("Neso","Nest"), Ade=c("Aeso","Aest")))