| Type: | Package |
| Title: | Replicability Analysis of High-Throughput Experiments |
| Version: | 1.0.2 |
| Description: | Implementing a computationally scalable false discovery rate control procedure for replicability analysis based on maximum of p-values. Please cite the manuscript corresponding to this package [Lyu, P. et al., (2023), <doi:10.1093/bioinformatics/btad366>]. |
| License: | GPL-3 |
| Encoding: | UTF-8 |
| Depends: | R (≥ 4.1.2), Rcpp, splines, stats |
| LinkingTo: | Rcpp, RcppArmadillo |
| RoxygenNote: | 7.2.3 |
| NeedsCompilation: | yes |
| Packaged: | 2025-05-30 12:28:20 UTC; P53 |
| Author: | Pengfei Lyu [aut, ctb], Yan Li [aut, cre, cph], Xiaoquan Wen [aut], Hongyuan Cao [aut, ctb] |
| Maintainer: | Yan Li <yanli_@jlu.edu.cn> |
| Repository: | CRAN |
| Date/Publication: | 2025-05-30 12:40:02 UTC |
Replicability Analysis of High-Throughput Experiments
Description
Replicability Analysis of High-Throughput Experiments
Usage
JUMP(pvals1, pvals2, alpha = 0.05, lambda = seq(0.01, 0.8, 0.01))Arguments
pvals1 | A numeric vector of p-values from study 1. |
pvals2 | A numeric vector of p-values from study 2. |
alpha | The FDR level to control, default is 0.05. |
lambda | The values of the tuning parameter to estimate pi_0. Must be in [0,1), default is seq(0.01, 0.8, 0.01). |
Value
a list with the following elements:
p.max | The maximum of p-values across two studies. |
jump.thr | The estimated threshold of p.max to control FDR at level alpha. |
Examples
# Simulate p-values in two studiesm = 10000h = sample(0:3, m, replace = TRUE, prob = c(0.9, 0.025, 0.025, 0.05))states1 = rep(0, m); states2 = rep(0, m)states1[which(h==2|h==3)] = 1; states2[which(h==1|h==3)] = 1z1 = rnorm(m, states1*2, 1)z2 = rnorm(m, states2*3, 1)p1 = 1 - pnorm(z1); p2 = 1 - pnorm(z2)# Run JUMP to identify replicable signalsres.jump = JUMP(p1, p2, alpha = 0.05)sig.idx = which(res.jump$p.max <= res.jump$jump.thr)Estimate threshold of maximum p-values across two studies to control FDR.
Description
Estimate threshold of maximum p-values across two studies to control FDR.
Usage
jump_cutoff(pa_in, pb_in, xi_in, alpha_in)Arguments
pa_in | A numeric vector of p-values from study 1. |
pb_in | A numeric vector of p-values from study 2. |
xi_in | The estimates of proportions of three null components. |
alpha_in | The FDR level to control, default is 0.05. |
Value
A list including the maximum of p-values and estimated threshold for FDR control.