4.R

R: How to make professional and beautiful plots

1) lnstall R and R Studio

2) Reference cards for data analysis

3) R packages

How to install R Packages:

> install.packages("PACKAGE")

# using bioconductor.org
> source("https://bioconductor.org/biocLite.R")
> biocLite("PACKAGE")

Common used packages

To load data:

RMySQL: read in data from a database or Excel.
xlsx: read in data from a Excel.

To manipulate data:

stringr: easy to learn tools for regular expressions and character strings.
reshape2: transform data between wide and long formats.

To visualize data:

gplots: heatmap.2
ggplot2: plotting system for R, based on the grammar of graphics
plotly: R package for creating interactive web-based graphs

To model data:

lme4, nlme: Linear and Non-linear mixed effects models
randomForest: Random forest methods from machine learning
glmnet: Lasso and elastic-net regression methods with cross validation
NMF: do non-negative matrix factorization

To do bioinformatics:

limma, DESeq2, edgeR: do differential expression: link
survival: Tools for survival analysis
motifRG, MotIV, MotifDb: motif search and enrichment
enrichR: GO term enrichment

To write your own R packages:

devtools: An essential suite of tools for turning your code into an R package.

To color your figure:

colorbrewer: Creates nice looking color palettes especially for thematic maps http://colorbrewer2.org/

4) Homework

Plot one heatmap using function "heatmap.2" in package "gplots" or package "pheatmap" . Here are some files to help you finish this job: plotHeatmap.zip .
Make your own plots using other packages, like plotly or ggvis.