prioritizeCandidates: Candidate gene prioritization
In palenic/genePrioritization: Find Potential Disease-Related Genes

Description Usage Arguments Value Author(s) References See Also Examples

View source: R/prioritizeCandidates.R

Finds potential disease-related genes based on their coexpression with known disease-related ("seed") genes. Takes as input a gene expression matrix, a list of seed genes and (optionally) a list of candidate genes; computes a ranked co-expression list for each seed gene; assigns to each candidate gene the product of its ranks within these lists as a total score; ranks them according to their increasing score. Lower scores indicate a higher coexpression, and thus a higher probability of being involved in the given phenotype. The function performs all steps of the analysis and returns a RankedGeneList object summarizing the results, but they can be performed manually by the user with the functions findCoexpression, rankGenes and candidateScoring.

1 2	prioritizeCandidates(counts, seedGenes, candidateGenes=NULL, antiCorrelation=FALSE)

`counts`	A gene expression matrix (genes in rows, samples in columns): can be a count matrix, a microarray result, as long as it's numeric. Rows have to be named with the gene symbol or ID. If desired, normalization has to be previously performed be the user.
`seedGenes`	A vector or list of seed gene symbols or IDs. If a list, each gene has to be a list element. At least one seed gene has to be present in the counts matrix rownames. Missing or duplicated genes are removed.
`candidateGenes`	(default NULL) A vector or list of candidate gene symbols or IDs. If a list, each gene has to be a list element. If NULL, all other genes in the matrix which are not part of the seed genes will be considered candidates. If supplied, least one candidate gene has to be present in the rank matrix rownames. Missing or duplicated genes are removed.
`antiCorrelation`	A logical value (default FALSE). If TRUE, anti-correlation will be considered as a significant correlation, thus strongly anti-correlated genes will have a high rank.

A RankedGeneList object

Chiara Paleni
Politecnico di Milano
Maintainer: Chiara Paleni
E-Mail: <chiara.paleni@polimi.it>

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2935433/
Piro, Rosario M et al. “Candidate gene prioritization based on spatially mapped gene expression: an application to XLMR.” Bioinformatics (Oxford, England) vol. 26,18 (2010): i618-24. doi:10.1093/bioinformatics/btq396

findCoexpression, rankGenes, candidateScoring, RankedGeneList

a <- matrix(c(1,2,3,2,4,6,8,6,4,5,2,8,7,1,5),
nrow=5, ncol=3,byrow=TRUE)
colnames(a) <- c('sample1','sample2','sample3')
rownames(a) <- c('gene1','gene2','gene3','gene4','gene5')
seed <- c('gene1')
candidates <- c('gene2','gene4')
z <- prioritizeCandidates(counts=a, seedGenes=seed,
candidateGenes=candidates, antiCorrelation=FALSE)