gu-mi

# Note: using the devel versions of both packages!
library(DESeq) # version 1.9.11
library(edgeR) # version 2.99.8
library(VennDiagram)

# Read in data ------------------------------------------------------------

## Use pasilla data
datafile = system.file( "extdata/pasilla_gene_counts.tsv", package="pasilla" )
datafile

gu-mi

## I came across a Chinese science blog which discusses an interesting stuff that girls should reject 
## the first 37% (for 1/e = 36.79%) pursuers in her life. A little R simulation as outlined in the 
## gist is shown below just for fun. In the snippet, m is the total number of boys you can choose. 
## In other words, they are potential husband for you. Larger number means high "quality" but these 
## numbers are randomly sampled, i.e., you don’t know whether the one who just walks into your life 
## is your Mr. Right or not. You reject the first m/e people (probably without any reason...), where 
## e is the Euler’s number, and choose from the rest of the m people. If you find any one of the 
## remaning boys better than the best boy you saw earlier in the first m/e ones, then you choose that 
## guy; otherwise, you have to make do with the last guy for the rest of your life...

gu-mi

# A simple approach to visually-weighted regression plots

doInstall <- TRUE  # Change to FALSE if you don't want packages installed.
toInstall <- c("ggplot2", "reshape2", "MASS")
if(doInstall){install.packages(toInstall, repos = "http://cran.us.r-project.org")}
lapply(toInstall, library, character.only = TRUE)

# Generate some data:
nn <- 1000
myData <- data.frame(X = rnorm(nn),

gu-mi

# Simple ggplot2 heatmap
# with colorBrewer "spectral" palette

doInstall <- TRUE  # Change to FALSE if you don't want packages installed.
toInstall <- c("ggplot2", "reshape2", "RColorBrewer")
if(doInstall){install.packages(toInstall, repos = "http://cran.us.r-project.org")}
lapply(toInstall, library, character.only = TRUE)

# Generate a random matrix
# This can be any type of numeric matrix,

gu-mi

# Randomly allocating observations into groups, for, e.g. cross-validation
kk <- 10  # Number of partitions, as in "kk-fold cross-validation."

# Here is a data.frame full of good data:
nn <- 1003
myData <- data.frame(matrix(rnorm(nn * 3), ncol = 3))
colnames(myData) <- LETTERS[1:3]

# This does not work:
whichK <- sample(LETTERS[1:kk], nrow(myData), replace = T)

gu-mi

#' Constructor
EmailClass <- function(name, email) {
	nc = list(
		name = name,
		email = email,
		get = function(x) nc[[x]],
		set = function(x, value) nc[[x]] <<- value,
		props = list(),
		history = list(),
		getHistory = function() return(nc$history),

gu-mi

# Simple ggplot2 heatmap, with optimal seriation

doInstall <- TRUE  # Change to FALSE if you don't want packages installed.
toInstall <- c("ggplot2", "reshape2", "RColorBrewer", "seriation")
if(doInstall){install.packages(toInstall, repos = "http://cran.us.r-project.org")}
lapply(toInstall, library, character.only = TRUE)

# Using U.S. Judge Rating Data
myData <- as.matrix(USJudgeRatings)

gu-mi

heatmap.3=function (x, Rowv = TRUE, Colv = if (symm) "Rowv" else TRUE,
    distfun = dist, hclustfun = hclust, dendrogram = c("both",
        "row", "column", "none"), symm = FALSE, scale = c("none",
        "row", "column"), na.rm = TRUE, revC = identical(Colv,
        "Rowv"), add.expr, breaks, symbreaks = max(x < 0, na.rm = TRUE) ||
        scale != "none", col = "heat.colors", colsep, rowsep,
    sepcolor = "white", sepwidth = c(0.05, 0.05), cellnote, notecex = 1,
    notecol = "cyan", na.color = par("bg"), trace = c("column",
        "row", "both", "none"), tracecol = "cyan", hline = median(breaks),
    vline = median(breaks), linecol = tracecol, margins = c(5,

gu-mi

# http://www.r-bloggers.com/example-10-7-fisher-vs-pearson/
# Author: Ken Kleinman

# In R, we'll simulate observations from a multinomial distribution 
# with the desired cell probabilities, and assemble the result into 
# a table to calculate the p-values. This will make it easier to simulate 
# tables under the alternative, as we need to do to assess power. 
# If there are empty rows or columns, the chisq.test() function produces 
# a p-value of "NaN", which will create problems later. To avoid this, 
# we'll put the table generation inside a while() function. This operates 

gu-mi

doInstall <- TRUE  # Change to FALSE if you don't want packages installed.
toInstall <- c("ReadImages", "reshape", "ggplot2")
if(doInstall){install.packages(toInstall, repos = "http://cran.r-project.org")}
lapply(toInstall, library, character.only = TRUE)

# Image URL:
allImageURLs <- c("http://media.charlesleifer.com/blog/photos/thumbnails/akira_940x700.jpg",
                  "http://upload.wikimedia.org/wikipedia/commons/thumb/e/ec/Mona_Lisa%2C_by_Leonardo_da_Vinci%2C_from_C2RMF_retouched.jpg/402px-Mona_Lisa%2C_by_Leonardo_da_Vinci%2C_from_C2RMF_retouched.jpg",
                  "http://upload.wikimedia.org/wikipedia/commons/thumb/e/e9/Official_portrait_of_Barack_Obama.jpg/441px-Official_portrait_of_Barack_Obama.jpg",
                  "http://cache.boston.com/universal/site_graphics/blogs/bigpicture/obama_11_05/obama22_16604051.jpg",