getPSS: Evaluate the CGNB score of metabolites

In MPINet: The package can implement the network-based metabolite pathway identification of pathways.

Description

Integrate the non-equivalence scores and the initial bias scores of metabolites by the monotonic spline model.

Usage

getPSS(riskmeta,plot=TRUE,binsize=400)

Arguments

riskmeta	A character vector of interesting metabolites, for each element is a pubchem CID.
plot	A logical. If TRUE the fit line obtained by the monotonic spline model will be plotted.
binsize	plot the fraction of differential metabolites in bins of this size.

Details

This function is used to integrate the non-equivalence of metabolites and the character of differential metabolites.The binsize can be assigned according to the best visualization.

Value

A data frame with 4 columns including "riskmeta", "meanstrvalue", "pss" and "CGNB". Each row correspond a metabolite pubchem CID. "riskmeta" indicates whether the metabolite is in the interesting set (with "1" is in and "0" is not in)."meanstrvalue" is the mean SOC value of the metabolite. "pss" is the score value obtained by the monotonic spline model. "CGNB" is the CGNB score of metabolite which is calculated as 1 substract the score value obtained by monotonic spline model. This score is used to calculate pathway weight in the subsequent pathway analysis.

Author(s)

Yanjun Xu <[e-mail]>, Chunquan Li <[e-mail]> and Xia Li <[e-mail]>

References

Young, M.D., Wakefield, M.J., Smyth, G.K. and Oshlack, A. (2010) Gene ontology analysis for RNA-seq: accounting for selection bias. Genome Biol, 11, R14.

Examples

## Not run: 

#######################################


##########get example data
risk<-GetExampleData(dataset="prostate")

###########calculate the CGNB score
pss<-getPSS(risk ,plot=F)
CGNBscore<-pss[,"CGNB"]
names(CGNBscore)<-rownames(pss)
##########print the CGNB score of some metabolites to screen
head(CGNBscore)
#identify dysregulated pathways
anncpdpre<-identifypathway(risk,pss,pathType="KEGG",method="MPINet",annlim=1,bglim=6)
#convert ann to data.frame
result<-printGraph(anncpdpre,pathType="KEGG",method="MPINet")
head(result)







## End(Not run)

Example output

Loading required package: BiasedUrn
Loading required package: mgcv
Loading required package: nlme
This is mgcv 1.8-28. For overview type 'help("mgcv-package")'.
 11953968   9548588    420804   5281997    440744    124148 
0.9956769 0.9956769 0.9956769 0.9956769 0.9956769 0.9956769 
your input componentList have 92 components in background
your input componentList have 85 components in network
   pathwayId                                         pathwayName
1 path:00330                     Arginine and proline metabolism
2 path:00232                                 Caffeine metabolism
3 path:00380                               Tryptophan metabolism
4 path:01040             Biosynthesis of unsaturated fatty acids
5 path:00120                      Primary bile acid biosynthesis
6 path:00130 Ubiquinone and other terpenoid-quinone biosynthesis
  annComponentRatio annBgRatio      weight       pvalue          fdr
1             10/92    89/4994 0.220476179 3.528762e-12 2.117257e-10
2              3/92    21/4994 0.005586311 1.457150e-09 4.371451e-08
3              3/92    80/4994 0.002731923 1.089882e-08 2.179764e-07
4              3/92    49/4994 0.005168241 1.622806e-08 2.434209e-07
5              2/92    47/4994 0.001598522 9.615284e-07 1.153834e-05
6              2/92    74/4994 0.001180019 1.323514e-06 1.287317e-05
  annComponentinNetRatio
1                  10/85
2                   3/85
3                   2/85
4                   2/85
5                   2/85
6                   2/85

MPINet documentation built on May 1, 2019, 8:04 p.m.