cummeRbund-example-workflow.Rnw

%
%
%
%\VignetteIndexEntry{Sample cummeRbund workflow}
%\VignetteKeywords{cummeRbund,visualization,sequencing,cufflinks,cuffdiff}
%\VignettePackage{cummeRbund}
%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%\SweaveOpts{prefix.string=graphics/plot}
\documentclass[10pt]{article}
\usepackage{amsmath}
\usepackage[authoryear,round]{natbib}
\usepackage{hyperref}
\hypersetup{
    colorlinks,
    citecolor=black,
    filecolor=black,
    linkcolor=red,
    urlcolor=black
}
\usepackage{theorem}
\usepackage{float}
\usepackage{ifthen}
\usepackage[OT1]{fontenc}

\newcommand{\R}{{\textsf{R}}}
\newcommand{\code}[1]{{\texttt{#1}}}
\newcommand{\term}[1]{{\emph{#1}}}
\newcommand{\Rpackage}[1]{\textsf{#1}}
\newcommand{\Rfunction}[1]{\texttt{#1}}
\newcommand{\Robject}[1]{\texttt{#1}}
\newcommand{\Rclass}[1]{{\textit{#1}}}
\newcommand{\Rmethod}[1]{{\textit{#1}}}
\newcommand{\Rfunarg}[1]{{\textit{#1}}}

\bibliographystyle{plainnat}
\title{cummeRbund: Sample Workflow}

\author{Loyal A. Goff, Cole Trapnell}
\date{}

\title{CummeRbund workflow}
\author{Loyal A. Goff}

\begin{document}

<<init, echo=FALSE>>=
options(width=65)
@ 

\maketitle

This document is a work in progress and will continually be updated as new features or analyses are integrated into the cummeRbund pipeline.
This guide is being released as is, with the understanding that existing gaps will be completed in due time.  Please bear with us as we work to expand this resource.

\tableofcontents

\section{Overview}

\section{Workflow Summary}

\section{Reading cuffdiff output}
<<loadLib>>=
library(cummeRbund)
@
<<read>>=
cuff <- readCufflinks(dir=system.file("extdata", package="cummeRbund"))
cuff
@
This example uses the 'test' dataset that is included in the cummeRbund package. Importantly, if your current working directory contains the output
from your cuffdiff analysis, simply calling \Rfunction{cuff<-readCufflinks()} will automatically find the default files and begin parsing your data.

\section{Quality Assessment of data}

\subsection{Evaluating model fit}
<<model_fit_1,include=FALSE>>=
d<-dispersionPlot(genes(cuff))
d
@

<<label=model_fit_1_plot,fig=TRUE,echo=FALSE>>=
<<model_fit_1>>
print(d)
@

\subsection{Identifying outlier replicates}
<<rep_boxplot_1,include=FALSE>>=
pBoxRep<-csBoxplot(genes(cuff),replicates=T)
pBoxRep
@
<<rep_dendro_1,include=FALSE>>=
pDendro<-csDendro(genes(cuff),replicates=T)
pDendro
@

<<label=rep_boxplot_1_plot,fig=TRUE,echo=FALSE>>=
<<rep_boxplot_1>>
print(pBoxRep)
@

<<label=rep_dendro_1_plot,fig=TRUE,echo=FALSE>>=
<<rep_dendro_1>>
print(pDendro)
@

\subsection{Determining relationships between conditions}
<<boxplot_1,include=FALSE>>=
pBox<-csBoxplot(genes(cuff))
pBox
@

<<label=boxplot_1_plot,fig=TRUE,echo=FALSE>>=
<<boxplot_1>>
print(pBox)
@


\section{Analysis of differential expression}

\subsection{Identifying differentially expressed genes}
<<diff_exp_genes_1>>=
sigGeneIds<-getSig(cuff,alpha=0.05,level="genes")
head(sigGeneIds)
length(sigGeneIds)
@

<<diff_exp_genes_2>>=
hESCvsFibroblast.sigGeneIds<-getSig(cuff,"hESC","Fibroblasts",alpha=0.05,level="genes")
head(hESCvsFibroblast.sigGeneIds)
length(hESCvsFibroblast.sigGeneIds)
@


\subsubsection{Creating significant gene sets}
<<diff_exp_genes_3>>=
sigGenes<-getGenes(cuff,sigGeneIds)
sigGenes
@

\subsubsection{Visualization of significant gene sets}

\subsection{Identifying differentially expressed features}
<<diff_exp_feat_1>>=
sigGeneIds<-getSig(cuff,alpha=0.05,level="isoforms")
head(sigGeneIds)
length(sigGeneIds)
@


\subsubsection{Creating significant feature sets}

\subsubsection{Visualization of significant feature sets}

\subsection{Analysis of individual genes}

\subsubsection{Subsetting individual genes}

\subsubsection{Creating a CuffGene object}
<<ind_gene_1>>=
Pink1<-getGene(cuff,'PINK1')
Pink1
@

\subsubsection{Visualization of individual genes}

\section{Data Exploration}

\subsection{Identifying patterns of gene expression}
Partitioning around medioids as e.g.

\subsection{Similarity analysis}
\Rmethod{findSimilar}

\subsection{Specificity analysis}
\Rmethod{csSpecificity}

\section{Analysis of differential splicing}

\section{Analysis of differential promoter usage}

\section{Linking with additional analyses}

\subsection{Gene Ontology Analysis (clusterProfiler)}
In this particular example, the reference .gtf file that was used has the official gene symbol (HUGO) stored in the \emph{gene name} attribute field. This value is carried over by cuffdiff
and cummeRbund into the \emph{gene\_short\_name} field for each feature.  We will exploit this ID and use it to map genes to their \emph{Entrez ID} to be able to use the very nice \Rpackage{clusterProfiler} package.

The first step is to translate

\subsection{Gene Set Enrichment Analysis (Preranked)}
We will generate .rnk and .gmx files for use in the GSEA Preranked analysis.

\subsection{Principal component analysis}

\section{Visualization Tips/Tricks}

\subsection{Faceting}


\section{Session Info}

\end{document}