Difference between revisions of "GGB-correlation-snippet.R"

From Organic Design wiki
m (fix k param)
(Fixing/moving plots)
Line 12: Line 12:
  
 
params <- c(2.74886, 1.36546, 4.12844) # "IPTG-a parameters"
 
params <- c(2.74886, 1.36546, 4.12844) # "IPTG-a parameters"
k <- 100
+
k <- 1000
  
 
a.shape  <- params[1]
 
a.shape  <- params[1]
Line 19: Line 19:
  
 
kDE <- rgamma(2*m1, shape=k, rate=k) # p49 2001 NEWTON paper
 
kDE <- rgamma(2*m1, shape=k, rate=k) # p49 2001 NEWTON paper
plot(kDE)
 
plot(density(kDE))
 
 
DEscales <- rgamma(2*m1, shape=a0.shape, rate=scale)
 
DEscales <- rgamma(2*m1, shape=a0.shape, rate=scale)
 
DEscales <- DEscales * kDE
 
DEscales <- DEscales * kDE
  
 
kEE <- rgamma(m0, shape=k, rate=k)
 
kEE <- rgamma(m0, shape=k, rate=k)
plot(kEE)
 
plot(density(kEE))
 
 
EEscales <- rgamma(m0 , shape=a0.shape, rate=scale)
 
EEscales <- rgamma(m0 , shape=a0.shape, rate=scale)
 
EEscales <- EEscales / kEE
 
EEscales <- EEscales / kEE
Line 37: Line 33:
  
 
# Graphical check
 
# Graphical check
 +
hist(kDE, prob=TRUE)
 +
lines(density(kDE))
 +
 +
hist(kEE, prob=TRUE)
 +
lines(density(kEE))
 +
 +
Xbar <- rowMeans(log2(X))
 +
hist(Xbar, breaks=30, prob=TRUE)
 +
lines(density(Xbar))
 
pairs(log2(X))
 
pairs(log2(X))
hist(rowMeans(log2(X)), breaks=30)
 

Revision as of 23:58, 28 August 2006

  1. p (genes) by n (slides) matrix
  2. m=p keeping fdr notation

m <- p <- 1000 nreps <- 5

n <- nreps * 2 pi0 <- 0.95

m1 <- round(m * (1-pi0)) m0 <- p - m1

params <- c(2.74886, 1.36546, 4.12844) # "IPTG-a parameters" k <- 1000

a.shape <- params[1] a0.shape <- params[2] scale <- params[3]

kDE <- rgamma(2*m1, shape=k, rate=k) # p49 2001 NEWTON paper DEscales <- rgamma(2*m1, shape=a0.shape, rate=scale) DEscales <- DEscales * kDE

kEE <- rgamma(m0, shape=k, rate=k) EEscales <- rgamma(m0 , shape=a0.shape, rate=scale) EEscales <- EEscales / kEE

scales <- c(rep(DEscales, each=nreps), rep(EEscales, each=2*nreps)) X <- rgamma(n* p, a.shape, rate=scales) dim(X) <- c(n,p)

X <- t(X)

  1. Graphical check

hist(kDE, prob=TRUE) lines(density(kDE))

hist(kEE, prob=TRUE) lines(density(kEE))

Xbar <- rowMeans(log2(X)) hist(Xbar, breaks=30, prob=TRUE) lines(density(Xbar)) pairs(log2(X))