% \VignetteEngine{knitr::knitr}
% \VignetteIndexEntry{07. Annotation and Visualization -- slides}

\documentclass{beamer}

\usepackage{BioconductorSlides}

\title{Annotation}
\author{Martin T.\ Morgan\footnote{\url{mtmorgan@fhcrc.org}}}
\date{27-28 February 2014}

<<knitr-settings, echo=FALSE>>=
library(knitr)
opts_chunk$set(tidy=FALSE, cache=TRUE)
@ 

\begin{document}

\maketitle

\begin{frame}{What is `Annotation'?}
  \begin{itemize}
  \item Genes -- classification schemes (e.g., Entrez, Ensembl),
    pathway membership, \ldots
  \item Genomes -- reference genomes; exons, transcripts, coding
    sequence; coding consequences
  \item System / network biology -- pathways, biochemical reactions,
    \ldots
  \end{itemize}
  Other defintions (not covered here): assigning function to novel
  sequence assemblies, \ldots
\end{frame}

\begin{frame}[fragile]{\Bioconductor{} Annotation Resources -- Packages}
  Model organism annotation packages
  \begin{itemize}
  \item \Rpackage{org.*} -- gene names and pathways
  \item \Rpackage{TxDb.*} -- gene models
  \item \Rpackage{BSgenome.*} -- whole-genome sequences
  \end{itemize}
\end{frame}

\section{Gene and pathway annotations}

\begin{frame}[fragile]{\Rpackage{org.*} packages}
  The `select' interface:
  \begin{itemize}
  \item Discovery: \Rfunction{keytypes}, \Rfunction{columns}, \Rfunction{keys}
  \item Retrieval: \Rfunction{select}
  \end{itemize}
<<select, eval=FALSE>>=
library(org.Hs.eg.db)
keytypes(org.Hs.eg.db)
columns(org.Hs.eg.db)
egid <- 
  select(org.Hs.eg.db, "BRCA1", "ENTREZID", "SYMBOL")
@ 
\end{frame}

\begin{frame}{\Rpackage{org.*} packages -- Useful \R{} commands}
  Within-\Rclass{vector} or \Rclass{data.frame}
  \begin{itemize}
  \item Finding and removing duplicates: \Rfunction{duplicated}, \Rfunction{unique}
  \item \Rfunction{any}, \Rfunction{all}
  \end{itemize}
  Between-\Rclass{vector} or \Rclass{data.frame}
  \begin{itemize}
  \item Matching \Rfunction{\%in\%}, \Rfunction{match}
  \item Set operations: \Rfunction{setdiff}, \Rfunction{union}, \Rfunction{intersect}
  \item \Rfunction{merge} Join two \Rclass{data.frame}s based on shared column.
  \end{itemize}
\end{frame}

\begin{frame}{\Rpackage{org.*} pacakges -- Under the hood\ldots}
  SQL (sqlite) data bases
  \begin{itemize}
  \item \Rcode{org.Hs.eg\_dbconn()} to query using \Rpkg{RSQLite}
    package
  \item \Rcode{org.Hs.eg\_dbfile()} to discover location and query
    outside \R{}.
  \end{itemize}
\end{frame}

\section{Genomes and genome coordinates}

\begin{frame}[fragile]{\Rpackage{TxDb.*} packages}
  \begin{itemize}
  \item Gene models for common model organsisms / genome builds /
    known gene schemes
  \item Supports the `select' interface (\Rfunction{keytypes},
    \Rfunction{columns}, \Rfunction{keys}, \Rfunction{select})
  \item `Easy' to build custom packages when gene model exist
  \end{itemize}
  Retrieving genomic ranges
  \begin{itemize}
  \item \Rfunction{transcripts}, \Rfunction{exons}, \Rfunction{cds},
  \item \Rfunction{transcriptsBy} , \Rfunction{exonsBy},
    \Rfunction{cdsBy} -- group by gene, transcirpt, etc.
  \end{itemize}
<<cdsBy, eval=FALSE>>=
library(TxDb.Hsapiens.UCSC.hg19.knownGene)
txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene
cdsByTx <- cdsBy(txdb, "tx")
@ 
\end{frame}

\begin{frame}[fragile]{\Rpackage{BSgenome.*} packages}
  Whole-genome sequences
  \begin{itemize}
  \item `Masks' when available, e.g., repeat regions
  \item Load chromosomes, range-based queries: \Rfunction{getSeq},
    \Rfunction{extactTranscriptsFromGenome}
  \end{itemize}
<<extactTranscriptsFromGenome, eval=FALSE>>=
library(BSgenome.Hsapiens.UCSC.hg19)
library(GenomicFeatures)
dna <- extractTranscriptsFromGenome(Hsapiens, cdsByTx)
@ 
\end{frame}

\section{Web resources}

\begin{frame}[fragile]{\Bioconductor{} Annotation Resources -- Web-based}
  Rich web resources
  \begin{itemize}
  \item \Biocpkg{biomaRt} (\url{http://biomart.org}),
    \Biocpkg{rtracklayer} (UCSC genome browser)
  \item \Biocpkg{ArrayExpress}, \Biocpkg{GEOquery}, Biocpkg{SRAdb}
  \item \Biocpkg{PSICQUIC}, \Biocpkg{KEGGREST}, \Biocpkg{uniprot.ws}, \ldots
  \item \Biocpkg{AnnotationHub}
  \end{itemize}
\end{frame}

\begin{frame}[fragile]{\Biocpkg{biomaRt}}
  \begin{itemize}
  \item \url{http://biomart.org}
  \item Drill-down discovery: \Rfunction{listMarts}, \Rfunction{listDatasets},
    \Rfunction{listFilters}, \Rfunction{listAttributes}
  \item Retrieval: \Rfunction{getBM}
  \end{itemize}
<<biomart, eval=FALSE>>=
library(biomaRt)
ensembl <-                      ## discover & use
   useMart("ensembl", dataset="hsapiens_gene_ensembl")
head(listFilters(ensembl), 3)
myFilter <- "chromosome_name"
myValues <- c("21", "22")
myAttributes <- c("ensembl_gene_id","chromosome_name")
res <-
    getBM(attributes=myAttributes, filters=myFilter,
          values=myValues, mart=ensembl)
@   
\end{frame}

\begin{frame}[fragile]{\Biocpkg{PSICQUIC}}
  \begin{itemize}
  \item \textbf{P}rotemics \textbf{S}tandard \textbf{I}nitiative
    \textbf{C}ommon \textbf{QU}ery \textbf{I}nterfa\textbf{C}e
  \item Programmatic access to molecular interaction data bases.
  \item \url{https://code.google.com/p/psicquic/}
  \end{itemize}
<<psicquic, eval=FALSE>>=
library(PSICQUIC)
## Query web service for available providers
psicquic <- PSICQUIC()
providers(psicquic)         # 25 available providers
## interactions between TP53 and MYC
tbl <- 
    interactions(psicquic, c("TP53", "MYC"), "9606")
nrow(tbl)                   # 7 interactions
@ 
%% 
See the package
\href{http://bioconductor.org/packages/release/bioc/vignettes/PSICQUIC/inst/doc/PSICQUIC.pdf}{vignette}
for additional detail.
\end{frame}

\begin{frame}[fragile]{\Biocpkg{AnnotationHub}}
  \begin{itemize}
  \item Large-scale genome resources, lightly curated for easy access
    from \R.
  \item Supports tab-completion, \Rfunction{metadata} discovery,
    selection and filtering.
  \end{itemize}
<<AnnotationHub, eval=FALSE>>=
library(AnnotationHub)
hub <- AnnotationHub()
hub        ## 10511 resources
@ 
\end{frame}

\section{Conclusions}

\begin{frame}{Conclusions}
  Rich annotation resources
  \begin{itemize}
  \item Model organism and custom \Rpackage{org.*}, \Rpackage{TxDb.*},
    \Rpackage{BSgenome.*} packages
  \item Web-based access to public (e.g., \Biocpkg{biomaRt} and
    \Bioconductor-specific (e.g., \Biocpkg{AnnotationHub}) resources
  \end{itemize}
\end{frame}

\end{document}