LPE (version 1.46.0)

adjBaseOlig.error.step1: Evaluates LPE variance function of M for quantiles of A within and experimental condition by dividing the A in 100 intervals.

Description

Genes are placed in bins/quantiles according to their average expression intensity. The function adjBaseOlig.error calculates a pooled variance of M for genes within these bins/quantiles of A for the replicates of the experimental condition contained in y. Here the assumption is that variance of the genes in each interval is similar.

Usage

adjBaseOlig.error.step1(y, stats=median, setMax=FALSE, q=0.01, df=10)

Arguments

y

y is a preprocessed matrix or data frame of expression intensities in which columns are expression intensities for a particular experimental condition and rows are genes.

stats

It determines whether mean or median is to be used for the replicates

setMax

If T then all variances below the max variance in the ordered distribution of variances are set to the maximum variance. If F then variances are left as is (recommended)

q

q is the quantile width; q=0.01 corresponds to 100 quantiles i.e. percentiles. Bins/quantiles have equal number of genes and are split according to the average intensity A.

df

df stands for degrees of freedom. It is used in smooth.spline function to interpolate the variances of all genes. Default value is 10.

Value

• Returns object of class baseOlig, comprising a data frame with 2 columns: A and var M. The A column contains the median values of each gene and the M columns contains the corresponding variance. Number of rows of the data-frame is same as that of the number of genes.

References

J.K. Lee and M.O.Connell(2003). An S-Plus library for the analysis of differential expression. In The Analysis of Gene Expression Data: Methods and Software. Edited by G. Parmigiani, ES Garrett, RA Irizarry ad SL Zegar. Springer, NewYork.

Jain et. al. (2003) Local pooled error test for identifying differentially expressed genes with a small number of replicated microarrays, Bioinformatics, 1945-1951.

Jain et. al. (2005) Rank-invariant resampling based estimation of false discovery rate for analysis of small sample microarray data, BMC Bioinformatics, Vol 6, 187.

See Also

lpeAdj

Examples

# Loading the data from the LPE library
  data(Ley)
  
  dim(Ley)
  # Gives 12488 by 7
  Ley[1:3,]
  # Returns 
  #       ID           c1   c2   c3    t1    t2    t3
  # 1  AFFX-MurIL2_at 4.06 3.82 4.28 11.47 11.54 11.34
  # 2 AFFX-MurIL10_at 4.56 2.79 4.83  4.25  3.72  2.94
  # 3  AFFX-MurIL4_at 5.14 4.10 4.59  4.67  4.71  4.67

  Ley[1:1000,2:7] <- preprocess(Ley[1:1000,2:7],data.type="MAS5")
  # Finding the baseline distribution of subset of the data
  # condition one (3 replicates)
  var.1 <- adjBaseOlig.error.step1(Ley[1:1000,2:4], setMax=FALSE, q=0.01)
  dim(var.1)
  # Returns a matrix of 1000 by 2 (A,M) format