| Title: | Spatial and intensity based normalization of cDNA microarray data based on robust neural nets |
|---|---|
| Description: | This package allows to detect and correct for spatial and intensity biases with two-channel microarray data. The normalization method implemented in this package is based on robust neural networks fitting. |
| Authors: | Adi Laurentiu Tarca <[email protected]> |
| Maintainer: | Adi Laurentiu Tarca <[email protected]> |
| License: | LGPL |
| Version: | 2.71.0 |
| Built: | 2024-10-30 09:02:32 UTC |
| Source: | https://github.com/bioc/nnNorm |
This function was concieved to easily compare several normalization methods in terms of variability of log-ratios, M. Basically it produces two plots: The first is a the density plot of the several matrices passed as arguments, while the second is a box plot. Median of absolute deviations for each method is printed on screen.
compNorm(x,...,bw="AUTO",xlim=c(-3,3),titles="AUTO",type="d")compNorm(x,...,bw="AUTO",xlim=c(-3,3),titles="AUTO",type="d")
x |
A vector of numerical values, e.q. the |
... |
An undefined number of objects similar with |
bw |
Band width required to compute the density distribution. |
xlim |
The range for abscissa of the density plots. |
titles |
Names to be displayed the charts legend. |
type |
If set to |
This function is used to compare the normalized log ratios obtained with several normalization methods.
NULL, this function only displays charts and prints on the screen some statistics.
Tarca, A.L.
A. L. Tarca, J. E. K. Cooke, and J. Mackay. Robust neural networks approach for spatial and
intensity dependent normalization of cDNA data. Bioinformatics. 2004,submitted.
# Normalize swirl data with two methods data(swirl) swirlNN<-maNormNN(swirl[,1]) swirlLoess<-maNormMain(swirl[,1]) nms<-c("None","Loess","NNets") #compare distributions: density plot compNorm(as.vector(maM(swirl[,1])),as.vector(maM(swirlLoess)),as.vector(maM(swirlNN)),xlim=c(- 2,2),bw="AUTO",titles=nms,type="d") #compare distributions: box plot compNorm(as.vector(maM(swirl[,1])),as.vector(maM(swirlLoess)),as.vector(maM(swirlNN)),xlim=c(- 2,2),bw="AUTO",titles=nms,type="b")# Normalize swirl data with two methods data(swirl) swirlNN<-maNormNN(swirl[,1]) swirlLoess<-maNormMain(swirl[,1]) nms<-c("None","Loess","NNets") #compare distributions: density plot compNorm(as.vector(maM(swirl[,1])),as.vector(maM(swirlLoess)),as.vector(maM(swirlNN)),xlim=c(- 2,2),bw="AUTO",titles=nms,type="d") #compare distributions: box plot compNorm(as.vector(maM(swirl[,1])),as.vector(maM(swirlLoess)),as.vector(maM(swirlNN)),xlim=c(- 2,2),bw="AUTO",titles=nms,type="b")
This function allows to identify in two channel batch of arrays, which are the print-tips where spatial bias is present.
detectSpatialBias(mbatch, corThreshold=0.6)detectSpatialBias(mbatch, corThreshold=0.6)
mbatch |
A |
corThreshold |
The correlation treshold to be used. |
This function computes two matrices: biasRow and biasCol. The elements of these
matrices represent the fraction of rows (columns) for which the correlation coefficient between log-ratios, M, and
column index (row index) is higher than a user specified treshold (default corThreshold=0.6).
The idea here is to see in which print-tip a important fraction of the rows (columns) are highly correlated with the
column (row) index. Since some rows (columns) will show positive correlation while the other negative correlation, we
are only interested in a sigle direction of the correlation, i.e. either positive or negative.
This function returns a list with two matrices. biasRow and biasCol. The rows of these matrices correspond
to the print tips counted metaRow wise, and the columns correspond to arrays. Values in these matrices superior to 33
point to print-tips that have more tha a third of the rows (columns) with important spatial bias.
Tarca, A.L.
A robust neural networks approach for spatial and intensity dependent normalization of cDNA microarray data,
Adi. L. Tarca , Janice. E. K. Cooke, and John Mackay, Bioinformatics, 21, 2005, 2674 - 2683.
# detecting spatial bias in swirl data data(swirl) # print-tip, intensity and spatial normalization of the first slide in swirl data set myres<-detectSpatialBias(swirl)# detecting spatial bias in swirl data data(swirl) # print-tip, intensity and spatial normalization of the first slide in swirl data set myres<-detectSpatialBias(swirl)
This function normalizes a batch of cDNA arrays by removing the intensity and spatial dependent bias.
maNormNN(mbatch,w=NULL,binWidth=3,binHeight=3,model.nonlins=3,iterations=100,nFolds=10,maplots=FALSE,verbose=FALSE)maNormNN(mbatch,w=NULL,binWidth=3,binHeight=3,model.nonlins=3,iterations=100,nFolds=10,maplots=FALSE,verbose=FALSE)
mbatch |
A |
w |
Weights to be assigned to each spot. If provided, it should be a vector with the same length as maNspots(mbatch). |
binWidth |
Width of the bins in the |
binHeight |
Height of the bins in the |
model.nonlins |
Number of nodes in the hidden layer of the neural network model. |
iterations |
The number of iterations at which (if not converged) the training of the neural net will be stopped. |
nFolds |
Number of cross-validation folds. It represents the number of equal parts in which the data from a print tip is divided into: the model is trained on nFolds-1 parts and the bias is estimated for one part at the time. Higher values improve the results but increase the computation time. Ideal values are between 5 and 10. |
maplots |
If set to |
verbose |
If set to |
This function uses neural networks to model the bias in cDNA data sets.
A marrayNorm object containing the normalized log ratios. See marrayNorm
class for details
Tarca, A.L.
A. L. Tarca, J. E. K. Cooke, and J. Mackay. Robust neural networks approach for spatial and
intensity dependent normalization of cDNA data. Bioinformatics. 2004,submitted.
compNorm,nnet
# Normalization of swirl data data(swirl) # print-tip, intensity and spatial normalization of the first slide in swirl data set swirlNN<-maNormNN(swirl[,1]) #do not consider spatial variations, and display M-A plots before and after normalization swirlNN<-maNormNN(swirl[,1],binWidth=maNsc(swirl),binHeight=maNsr(swirl),maplots=TRUE)# Normalization of swirl data data(swirl) # print-tip, intensity and spatial normalization of the first slide in swirl data set swirlNN<-maNormNN(swirl[,1]) #do not consider spatial variations, and display M-A plots before and after normalization swirlNN<-maNormNN(swirl[,1],binWidth=maNsc(swirl),binHeight=maNsr(swirl),maplots=TRUE)