lp_norm_k_powers_surv {GSSTDA}

Filtering function

Description

A filtering function for mapper that projects $$R$^n$ into $R$. It calculates for each column of the matrix (each patient), its value of the filtering function. Specifically, it computes the vector magnitude in the Lp norm (as well as k powers of this magnitude) of the vector resulting of weighting each element of the column vector by the Z score obtained by fitting a cox proportional hazard model to the level of each gene. For further information see "Progression Analysis of Disease with Survival (PAD-S) by SurvMap identifies different prognostic subgroups of breast cancer in a large combined set of transcriptomics and methylation studies"

Usage

lp_norm_k_powers_surv(genes_disease_component, p, k, cox_all_matrix)

Arguments

genes_disease_component

Disease component matrix (output of the function of generate_disease_component), after having selected the rows corresponding to the selected genes.

p

integer. It indicates the p norm to be calculated. If k = 1 and p = 2, the function computes the standard (Euclidean) vector magnitude of each column. For larger values of p the weight of genes with larger levels is greater.

k

integer. Powers of the vector magnitude. If k = 1 and p = 2, the function computes the standard (Euclidean) vector magnitude of each column.

cox_all_matrix

A matrix with the output of the cox_all_genes function that stores the information of all cox proportional hazard model tests for each gene in the dataset.

Value

A numeric vector including the values produced by the function for each sample in the dataset.