lassoReg {a4Base} | R Documentation |
Multiple regression using the Lasso algorithm as implemented in the glmnet package. This is a theoretically nice approach to see which combination of genes predict best a continuous response. Empirical evidence that this actually works with high-dimensional data is however scarce.
lassoReg(object, covariate)
object |
object containing the expression measurements; currently the only method supported is one for ExpressionSet objects |
covariate |
character string indicating the column containing the continuous covariate. |
object of class glmnet
Willem Talloen
Goehlmann, H. and W. Talloen (2009). Gene Expression Studies Using Affymetrix Microarrays, Chapman \& Hall/CRC, pp. 211.
if (require(ALL)){ data(ALL, package = "ALL") ALL <- addGeneInfo(ALL) ALL$BTtype <- as.factor(substr(ALL$BT,0,1)) resultLasso <- lassoReg(object = ALL[1:100,], covariate = "age") plot(resultLasso, label = TRUE, main = "Lasso coefficients in relation to degree of penalization.") featResultLasso <- topTable(resultLasso, n = 15) }