cor_phenotype: Correlate inferred components against phenotype
In jinghan1018/tensorICA: tensorICA

Description Usage Arguments Value References Examples

View source: R/cor_phenotype.R

Correlate inferred components against phenotype.

1	cor_phenotype(tica.o, phenotype, phenotype.is.categorical, component = 0)

`tica.o`	the TICA result object, which can be obtained by function DoTICA.
`phenotype`	A list comtains phenotype vectors inputed by users to correlate against, each should be in the same length with the number of sample.
`phenotype.is.categorical`	A Boolean logical vector states whether the phenotype is categorical. It is of the same length of phenotype list.
`component`	Choose the component significantly correlated to specific phenotype. It is an optional parameter for boxplot or scatter plot of chosen component against phenotype.

A_star_matrix.p A heatmap of the second mode mixing matrix.

pv.p A heatmap of p value between phenotypes and the second mode mixing matrix components using linear regression model.

compheno.pl A list of boxplots or scatter plots of chosen component against all phenotype.

Teschendorff AE, Han J, Paul D, Virta J, Nordhausen K. Tensorial Blind Source Separation for Improved Analysis of Multi-Omic Data. Genome Biology (2018) 19:76.

data(buccalbloodtensor);
Dim.l <- EstDim(buccalbloodtensor$data);
dim <- Dim.l$dim;
tica.o <- DoTICA(Data = buccalbloodtensor$data, dim = dim, method = "FOBI");
phenotype.p <- cor_phenotype(tica.o = tica.o, phenotype = buccalbloodtensor$pheno.l,
 phenotype.is.categorical = buccalbloodtensor$pheno.i);
phenotype.p$A_star_matrix.p;
phenotype.p$pv.p;