PRODIGY: PRODIGY

In Shamir-Lab/PRODIGY: Personalized prioritization of driver genes

PRODIGY

Description

This function runs the PRODIGY algorithm for a single patient.

Usage

PRODIGY(
  mutated_genes,
  expression_matrix,
  network = NULL,
  sample,
  diff_genes = NULL,
  alpha = 0.05,
  pathway_list = NULL,
  num_of_cores = 1,
  sample_origins = NULL,
  write_results = F,
  results_folder = "./",
  beta = 2,
  gamma = 0.05,
  delta = 0.05
)

Arguments

mutated_genes	A vector of mutated genes to examine using PRODIGY.
expression_matrix	A read count matrix with genes in rows and patients on columns. All genes must be contained in the global PPI network.
network	The global PPI network. Columns describe the source protein, destination protein and interaction score respectively. The network is considered as undirected.
sample	The sample label as appears in the SNV and expression matrices.
diff_genes	A vector of the sample's differentially expressed genes (with gene names). All genes must be contained in the global PPI network.
alpha	the penalty exponent.
pathway_list	A list where each object is a 3 column data.table (src,dest,weight). Names correspond to pathway names
num_of_cores	The number of CPU cores to be used by the influence scores calculation step.
sample_origins	A vector that contains two optional values ("tumor","normal") corresponds to the tissues from which each column in expression_matrix was derived. This vector is utilized for differential expression analysis. If no vector is specified, the sample names of expression_matrix are assumed to be in TCGA format where last two digits correspond to sample type: "01"= solid tumor and "11"= normal.
write_results	Should the results be written to text files?
results_folder	Location for resulting influence matrices storage (if write_results = T)
beta	Minimal fold-change threshold for declering gene as differentially expressed by DESeq (default = 0.2)
gamma	FDR threshold for declering gene as differentially expressed by DESeq (default = 0.05)
delta	FDR threshold for declering a pathway as statistically enriched for differentially expressed genes (default = 0.05)

Value

A matrix of influence scores for every mutation and every enriched pathway.

References

Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 1-21 (2014). Gabriele Sales, Enrica Calura and Chiara Romualdi, graphite: GRAPH Interaction from pathway Topological Environment (2017). Gillespie, M., Vastrik, I., Eustachio, P. D., Schmidt, E. & Bono, B. De. Reactome: a knowledgebase of biological pathways. Nucleic Acids Res. 33, 428-432 (2005). Schaefer, C. F. et al. PID: The pathway interaction database. Nucleic Acids Res. 37, 674-679 (2009). Ogata, H. et al. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 27, 29-34 (1999).

Examples

# Load SNV+expression data from TCGA
data(COAD_SNV)
data(COAD_Expression)
# Load STRING network data 
data(STRING_network)
network = STRING_network
sample = intersect(colnames(expression_matrix),colnames(snv_matrix))[1]
# Identify sample origins (tumor or normal)
sample_origins = rep("tumor",ncol(expression_matrix))
sample_origins[substr(colnames(expression_matrix),nchar(colnames(expression_matrix)[1])-1,nchar(colnames(expression_matrix)[1]))=="11"] = "normal"	
res = PRODIGY<-function(snv_matrix,expression_matrix,network=network,sample,diff_genes=NULL,alpha=0.05,pathwayDB="reactome",num_of_cores=1,sample_origins = sample_origins,beta=2,gamma=0.05,delta=0.05)

Shamir-Lab/PRODIGY documentation built on March 27, 2022, 5:29 p.m.