pcor.confint {GeneNT} | R Documentation |
This function returns confidence intervals for a list of partial correlation coefficients given empirical sample size kappa and significant level alpha.
pcor.confint(pcor, kappa, alpha)
pcor |
Partial correlation coefficient(s) |
kappa |
Empirical sample size |
alpha |
The significant level |
For details about the empirical sample size, refer to Schäfer, J., and K. Strimmer, 2004, Bioinformatics, 1:1-13. Briefly, it refers sample size that are estimated from data using a two-component mixture model rather than from sample size in the situation of "small N, large p" in the Gaussian Graphic Model framework.
The function returns asymptotic lower and upper bound(s) of Pearson correlation coefficient(s).
conf.int1 |
Lower bound(s) of Pearson correlation coefficient(s) |
conf.int2 |
Upper bound(s) of Pearson correlation coefficient(s) |
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.
Schäfer, J., and K. Strimmer. (2004) An empirical Bayes approach to inferring large-scale gene association networks. Bioinformatics, 1, 1-13.
pcor <- runif(20, -1, 1) #simulate a vector of correlation coefficients #g7 <- pcor.confint(pcor, 20, 0.05) #g7$conf.int1 #g7$conf.int2