get_sign_pathways: Detects signature pathways related to two patient classes

In LucaGiudice/Simpati: Pathway-based classifier exploits patient similarity network paradigm

Description

Detects the pathways that are signature between two patient classes. It does not depend by a step of classification. It computes a patient similarity network per pathway and finds the PSNs which significantly separate the two classes. It tests if a signature pathway specific PSN is signficant based on a permutation approach.

Usage

get_sign_pathways(
  geno_p,
  pathways_l1,
  vars_list,
  th_pred = 3,
  n_test = 200,
  Pv_th = 0.05,
  n_cores = 2
)

Arguments

geno_p	Matrix of patient's profiles
pathways_l1	List of feature's sets
vars_list	Variable list of the Simpati's run created with the set_variables function
th_pred	Default 3, Power threshold to understand if Simpati should try to perform the Best Friend Connector algorithm
n_test	Default 200, number of permutations to determine the p.value of significancy
Pv_th	Default 0.05, value of threshold to assess a pathway-specific PSN as significant
n_cores	Default 2, value which sets the number of cores to use for the parallel computation

Value

Return PSN_info_df: matrix of information related to the tested pathway specific PSNs PSN_comp_l: list of vectorized pathway specific PSNs outlier_df: matrix which summerizes the likelihood for each patient to be outlier of its a priori own class

LucaGiudice/Simpati documentation built on Jan. 27, 2022, 11:42 p.m.