GuangchuangYu

mirror	speed
http://cran.stat.upd.edu.ph/	5.212
http://healthstat.snu.ac.kr/CRAN/	5.322
http://cran.ism.ac.jp/	5.327
http://ftp.yzu.edu.tw/CRAN/	5.340
http://ftp.iitm.ac.in/cran/	5.380
https://mirrors.eliteu.cn/CRAN/	5.398
http://cran.stat.nus.edu.sg/	5.445

GuangchuangYu

d <- data.frame(pkg = c("GOSim", "GOSemSim", "DOSE", "meshes",
                        "DOSim", "HPOSim", "ppiPre", "MeSHSim", "meshr"),
                year = c(2007, 2010, 2015, 2018, 2011, 2015, 2013, 2015, 2015),
                journal = c("BMC Bioinformatics", rep("Bioinformatics", 3),
                            "BMC Bioinformatics", "PLoS One", "BMC Systems Biology",
                            "Journal of Bioinformatics and computational biology",
                            "BMC Bioinformatics"),
                y = c(1, 1, 2, 3, 2, 4, 5, 2.8, 3.2))

require(ggplot2)

GuangchuangYu

## devtools::install_github("stephenturner/msigdf")
library(msigdf)
library(dplyr)
library(clusterProfiler)

c2 <- msigdf.human %>% 
    filter(collection == "c2") %>% select(geneset, entrez) %>% as.data.frame

data(geneList)
de <- names(geneList)[1:100]

GuangchuangYu

## Antithetic sampling reframes our estimate as a sum of negatively
## correlated random variables, using the fact that negative
## correlation reduces the variance of a sum.

## http://en.wikipedia.org/wiki/Antithetic_variates

g <- function(x) 1/(1+x)
N <- 1500
n <- 50
u1 <- matrix(runif(2 * n * N), ncol = n)