Lecture 4 Distance from Two Distributaries

1
Lecture 4 Distance from Two DistributariesR
packages
2
General suggestion

• Do not have to remember the formulaNot to say
  the exotic names of the formula ? !

3
Distance from two distributions
4
Distance computation

• xlt-array(110, c(5,2))
• dist(x, method"euclidean")
• More distance functions are available in the
  bioDist package

5
Installing a new software package

• R is supported by research community- I found
  everything I wanted
• Popular places for find R packages
• http//cran.r-project.org/src/contrib/PACKAGES.ht
  ml
• http//bioconductor.org/packages/1.8/Software.htm
  l

6
Installing the bioDist package

• Go to http//bioconductor.org/packages/1.8/Softwa
  re.html
• Download bioDist_1.4.0.zip
• In R menu, click Packages-gtInstall packages from
  local zip files choose the downloaded zip file.
• In R menu, click Packages-gtLoad Package choose
  the bioDist package

7
Using the bioDist package

• spearman.dist
• tau.dist
• KLdist.matrix
• mutualInfo
• xlt-array(110, c(2,5))
• tau.dist(x)
• mutualInfo(x)

8
Intro to BioConductor

• What is it?
• An open source software project
• A well collection of data analysis packages
• Installation
• source("http//www.bioconductor.org/biocLite.R")
• biocLite()
• Menu Packages-gt Load package
• library(package_name)

9
Dataset, preprocessing distances

• http//www.bioconductor.org/mogr/chapter-code/dist
  .R

10
Homework (Due Sept 11)

• Read the Golub paper and prepare for discussion
  after the in class presentation
• Install R and BioConductor
• Read Gentleman book 13.1-13.3
• (Team) Execute this programhttp//www.bioconduct
  or.org/mogr/chapter-code/dist.R(install all
  necessary libraries)
• (Optional) Read Gentleman book 12.4