findMultiplePeaks {MetaNetwork}R Documentation

Cluster multiple mass peaks based on correlation and mass

Description

Predict metabolites that may be represented by multiple mass peaks based on correlation and m/z value. If two mass peaks are highly correlated (r>0.95) and their mass difference is 1 or 2 (as isotopes) or mass ratio is 2, 3, or 1/2, 1/3 (as different charged), they are predicted as multiple peaks of a same metabolite.

Usage

findMultiplePeaks <- function( corrZeroOrder, peaks, corrThres=0.95, 
                               filename=NULL) )

Arguments

corrZeroOrder a zero order peak correlation matrix that has been calculated before.
See corrZeroOrder example data.
peaks matrix of mass/charge peaks (column1) for each trait (rownames). See peaks2 example peaks data for unidentified example traits data traits2.
corrThres (optional) numeric threshold for significant correlated peaks that should be tested for multiple peaks relationships. Default is 0.95.
filename (optional) path of the csv file where the multiple peak summary is to be stored. Default NULL.

Value

Returns a data frame with Multiple Peaks summary containing the following headers:

cluster cluster id of related multiple peaks.
peak1 mass over charge peak.
mz1 mass over charge ratio of peak1.
peak2 mass over charge peak that is related to peak1.
mz2 mass over charge ratio of peak2.
corrCoef correlation coefficient between peak1 and peak2.
massDiff peak difference: mz1 - mz2.
massRatio peak ratio: mz1/mz2.
relationship putative isotope or diffCharged relationship.

See multiplePeaks example data set.

Note

The names of traits should be consistent over corrZeroOrder and peaks.

Author(s)

Jingyuan Fu <j.fu@rug.nl>, Morris Swertz <m.a.swertz@rug.nl>, Ritsert Jansen <r.c.jansen@rug.nl>

References

Fu J, Swertz MA, Keurentjes JJB, Jansen RC. MetaNetwork: a computational tool for the genetic study of metabolism. Nature Protocols (2007).

http://gbic.biol.rug.nl/supplementary/2007/MetaNetwork

See Also

Use traits2 and peaks2 as example unidintified peak data set or use loadData to load your own data.
Use qtlCorrZeroOrder to calculate corrZeroOrder zero order correlation.

Examples

## load the example data provided with this package                         
data(traits2)   
data(peaks2)    
data(genotypes)     
data(markers)                                         
                                             
##OR: load your own data                     
#traits2      <- loadData("traits2.csv")
#peaks2       <- loadData("peaks2.csv")
#genotypes    <- loadData("genotypes.csv")
#markers      <- loadData("markers.csv")
                                             
##calculate the two part qtl
qtlProfiles   <- qtlMapTwoPart(genotypes=genotypes, traits=traits2, spike=4)
  
##set the qtl threshold
qtlThres      <- 3.79

##OR: estimate the threshold yourself
#qtlThres     <- qtlThreshold(genotypes, traits, spike=4)
  
##calculate zero order correlation
corrZeroOrder <- qtlCorrZeroOrder(markers, qtlProfiles, qtlThres)

##find multiple peaks
multiplePeaks <- findMultiplePeaks(corrZeroOrder, peaks2)

##show summary
multiplePeaks
[Package MetaNetwork version 1.0-0 Index]