SASPECT {SASPECT}R Documentation

Significant AnalysiS of PEptide CounTs

Description

A function for identifying differentially expressed proteins between two sample groups using spectral counts from LC-MS/MS Experiments

Usage

SASPECT(peptideData, pep.set, pep.pro.name, run.group.info, 
        permu.iter=50, filter.run=2, filter.score=0.95)

Arguments

peptideData a list of two components: PeptideCount and PeptideConfidence. Both are numeric matrices with p rows each representing one peptide and n1+n2 columns each representing one sample (n1=sample size of the first group, and n2=sample size of the second group). PeptideCount records the peptide spectral counts of all p peptides in all n1+n2 samples. PeptideConfidence tracks the confidence score of each peptide identification in the database search procedure (e.g. the PeptideProphet score). Both matrics need to be arranged in the way that the first n1 columns represents samples from the first group and the rest columns are for the second group.
pep.set a character vector of length p. The ith element is the peptide ID corresponding to the ith row of peptideData$PeptideCount and peptideData$PeptideConfidence.
pep.pro.name a character matrix with 2 columns. The first column gives the protein IDs, and the second column gives the names of the peptides matching to the proteins in the first column.
run.group.info a data frame with two columns. The first column (run.group.info$label) is a character vector of length 2, giving the group names of the two groups. The second column (run.group.info$count) is a numeric vector of length 2, giving the number of samples in the first group (n1) and the second group (n2).
permu.iter an integer. It is the number of permutation iterations for estimating FDR. The default value is 50.
filter.run an integer. It is the filter criteria for removing peptides observed in too few samples. The default value is 2.
filter.score a scale. PeptideConfidence scores above this value are counted in the filtering process. The default value is 0.95

Details

This function implements the SASPECT-hybrid method (Wang et. al. 2008, in preparation), which is a modified version of the original SASPECT mothod proposed in Whiteaker et. al. 2007. The Score1 column in the returned matrix gives test statistics using the original SASPECT method.

Value

SASPECT generates a data frame with 7 columns:

Protein Protein groups' ID.
ProteinsInGroup Names of proteins in each protein group (separated by .).
Score1 test score based on Appear-Absent (AA) measurements. A positive value suggests the abundence level in the second group is higher than the first group. A negative value suggests the opposite.
Score2 test score based on non zero total Spectral count (SpecC) measurements. A positive value suggests the abundence level in the second group is higher than the first group. A negative value suggests the opposite.
Score final SASPECT score (sum square of Score1 and Score2).
Qvalue FDR resulted from permutation test based on Score.
PeptideNumber number of peptides observed for each protein(protein group).

Author(s)

Wang, P. and Liu, Y.

References

Whiteaker, J. R., Zhang, H., Zhao, L., Wang, P., Kelly-Spratt, K. S., Ivey, R. G., Piening, B. D., Feng, L., Kasarda, E., Gurley, K. E., Eng, J. K., Chodosh, L. A., Kemp, C. J., McIntosh, M. W., Paulovich, A. G (2007) Integrated Pipeline for Mass Spectrometry-Based Discovery and Confirmation of Biomarkers Demonstrated in a Mouse Model of Breast Cancer. J. Proteome Res., 6(10); 3962-3975.

Wang, P., Liu, Y., McIntosh, M. W., Paulovich, A. G (2008) Significant analysis for comparative proteomics studies using label free LC-MS/MS experiments (in preparation).

Examples


library(SASPECT)
data(mouseTissue)

SASPECT.result<-SASPECT(peptideData=mouseTissue$peptideData, 
        pep.set=mouseTissue$pep.set, 
        pep.pro.name=mouseTissue$pep.pro.name, 
        run.group.info=mouseTissue$run.group.info,
        permu.iter=50,
        filter.run=2,
        filter.score=0.95)
### it takes about 1 minute to run this example. 

### check the qvalue distribution
qvalue=as.numeric(SASPECT.result[,"Qvalue"])
plot(sort(qvalue))
     
### output the result into a table file
write.table(SASPECT.result, file="SASPECT.result.txt", row.names=FALSE, sep="\t")
     

[Package SASPECT version 0.1-1 Index]