BEST.kendall {GeneNT} | R Documentation |
This function implements the BLAST type two-stage screening procedure based on Kendall correlation statistic. Specifying a pair of FDR and MAS criteria, and a seed gene, the algorithm screens a seeded gene cluster with controlled FDR and MAS.
BEST.kendall(gene.name, Q, cormin)
gene.name |
The "seed" gene name |
Q |
The significant level |
cormin |
The specified minimum acceptable strength of association measured using Kendall correlation statistic |
The data matrix file must be in the right format. The first row (header) must be one shorter than the rest rows. The first column must be gene names.
The function returns a seeded gene cluster that satisfies the FDR and MAS constraints simultaneously using Kendall correlation statistic.
bkG1 |
The gene pairs that passes Stage I (FDR only) screening |
bkG2 |
The gene pairs that passed both Stage I (FDR) and II (MAS) screenings |
Dongxiao Zhu (http://www-personal.umich.edu/~zhud)
Zhu, D., Hero, A.O., Qin, Z.S. and Swaroop, A. High throughput screening of co-expressed gene pairs with controlled False Discovery Rate (FDR) and Minimum Acceptable Strength (MAS). Submitted.
Hollander M. and Wolfe D.A. (1999). Nonparametric Statistical Methods, New York: Wiley.
# load GeneNT and GeneTS library library(GeneTS) library(GeneNT) #EITHER use the internal dataset data(dat) #OR use the following if you want to import external data #dat <- read.table("gal.txt", h = T, row.names = 1) #Note, data matrix name has to be "dat" #use (FDR, MAS) criteria (0.2, 0.5) and seed gene "GAL7" as example to screen gene pairs #g5 <- BEST.kendall("GAL7", 0.2, 0.5) #bkG1 <- g5$bkG1 #bkG2 contains gene pairs that passed the two-stage screening #bkG2 <- g5$bkG2