%
% NOTE -- ONLY EDIT THE .Rnw FILE!!!  The .tex file is
% likely to be overwritten.
%
\documentclass[landscape]{article}

\usepackage{amsmath,pstricks}
\usepackage[authoryear,round]{natbib}
\usepackage{hyperref}
\usepackage{sectsty}
\usepackage{landscape}
\usepackage{graphics}

\textwidth=10.9in
\textheight=6.5in
%\parskip=.3cm
\oddsidemargin=.0in
\evensidemargin=.0in
\headheight=-.3in


\newcommand{\scscst}{\scriptscriptstyle}
\newcommand{\scst}{\scriptstyle}
\newcommand{\dimfo}{\fontsize{14}{16}\selectfont}
\newcommand{\expfo}{\fontsize{18}{22}\selectfont}


\newcommand{\Rfunction}[1]{{\texttt{#1}}}
\newcommand{\Robject}[1]{{\texttt{#1}}}
\newcommand{\Rpackage}[1]{{\textit{#1}}}
\newcommand{\Rmethod}[1]{{\texttt{#1}}}
\newcommand{\Rfunarg}[1]{{\texttt{#1}}}
\newcommand{\Rclass}[1]{{\textit{#1}}}

\newcommand\bi{\begin{itemize}}
\newcommand\ei{\end{itemize}}

\textwidth=8.2in

\bibliographystyle{plainnat} 

\renewcommand{\familydefault}{\sfdefault}

\usepackage[T1]{fontenc}

\newcommand\cp{{\clearpage}}
\newcommand\sts[1]{\Huge \textbf{#1}}
\newcommand\stsh[1]{\huge \textbf{#1}}
\newcommand\stsL[1]{\Large \textbf{#1}}
\newcommand\stsl[1]{\large \textbf{#1}}

\begin{document}

\fontsize{18}{22}
\selectfont
\allsectionsfont{\sffamily}
\sectionfont{\fontfamily{phv}\fontsize{18}{22}\selectfont}
\subsectionfont{\fontfamily{phv}\fontsize{18}{22}\selectfont}
\subsubsectionfont{\fontfamily{phv}\fontsize{18}{22}\selectfont}


%\setkeys{Gin}{width=0.85\textwidth}

{
\Huge
\begin{center}
\textbf{Lecture 4: annotation in bioconductor \\ \copyright 2008 VJ Carey PhD \\
Channing Lab}
\end{center}
\begin{itemize}
\item commitments
\item platform annotations -- SQLite basis
\item organism annotation -- org.Hs.eg.db
\item web services: biomaRt
\ei
}

\clearpage
{\sts{Annotation concept review}
\bi
\item platforms: feature annotation
\item experiments: MIAME annotation
\item samples: phenotype, disease, protocol
\item genomes, biological processes: ontologies for sequence, gene products, etc.
\item resources: networks of databases and tables
\ei
}

\clearpage

{\sts{Bioconductor commitments}
\bi
\item identifier annotation maps for prevalent platforms are created every
three months
\item open tools for annotating custom platforms are provided;
advice given
\item large-scale annotation sets for important organisms are
also provided
\bi
\item org.Hs.eg.db
\ei
\item harder problem: feature maps (e.g., CDF files for affy) for chips
depend on manufacturer openness, tractability -- see pd.mapping packages
\ei
}

\clearpage

{\stsL{Platforms: Classic examples}
<<lkanno>>=
library(hgfocus.db)
objects("package:hgfocus.db")
library(annotate)
nn = ls(hgfocusSYMBOL)[1:3]
mget( nn, hgfocusSYMBOL )
@
}

\clearpage
{\stsL{Platforms: Classic examples}
<<lkcd>>=
library(hgfocuscdf)
hgfocuscdf
pid = ls(hgfocuscdf)[1:3]
lapply( mget(pid, hgfocuscdf), "[", 1:3, 1:2 )


library(affy)
indices2xy(62627, cdf="hgfocuscdf")
@
}

\clearpage
{\stsL{Platforms: Classic examples}
<<lkpr>>=
library(hgfocusprobe)
hgfocusprobe
as.data.frame(hgfocusprobe)[1:3,]
@
}

\clearpage
{\stsL{Platforms: new approach (oligo)}
<<doma>>=
library(pd.mapping50k.xba240)
pd.mapping50k.xba240
@
}

\clearpage
{\stsL{pd.mapping approach}
<<dodoi>>=
xb = pd.mapping50k.xba240@getdb()
dbListTables(xb)
dbGetQuery(xb, "select * from featureSet limit 200,3")[,-11]
@
}

\clearpage
{\stsL{illumina lumi twist}
<<dolu>>=
library(lumiHumanV2)
library(lumi)
kk = ls(lumiHumanV2REFSEQ)[1050:1054]
kk
nuID2targetID(kk, lib="lumiHumanV2")
id2seq(kk)
@
}


\clearpage
{\stsL{illumina expression arrays: SQLite approach}
<<dod>>=
library(illuminaHumanv2.db)
ilco = illuminaHumanv2_dbconn()
dbListTables(ilco)
@
}

\clearpage
{\stsL{creating a hyperlinked page of annotation resolutions}
<<doa>>=
library(annaffy)
melaFrag = c("EGF", "EGFR", "HRAS", "ARAF", "PIK3R5", "MAP2K1", "NRAS", 
   "MAPK1", "BRAF")
library(GSEABase)
gs = GeneSet(melaFrag, geneIdType=SymbolIdentifier())
geneIdType(gs) = AnnotationIdentifier("illuminaHumanv2.db")
imf = geneIds(gs)
tab = aafTableAnn(imf, "illuminaHumanv2.db")
saveHTML(tab, file="mftab.html")
@
}

\clearpage

{\sts{organism-level annotation}

\setkeys{Gin}{width=0.95\textwidth}
\includegraphics{orgShot}

}

\clearpage
{\stsL{queries}
<<dodHs, keep.source=TRUE>>=
library(org.Hs.eg.db)
oc = org.Hs.eg_dbconn()
dbListTables(oc)
dbGetQuery(oc, "select * from gene_info limit 199, 5")
dbGetQuery(oc, "select * from gene_info INNER JOIN go_MF
    using(_id) where gene_info.symbol = 'TBX21' ")
@
}

\clearpage

{\stsL{pathway and gene ontology resources}
<<dopa>>=
library(KEGG.db)
objects("package:KEGG.db")
library(GO.db)
objects("package:GO.db")
@
}

\clearpage

{\stsL{web services: biomaRt}
<<dom>>=
library(biomaRt)
allm = listMarts()
allm
@
}

\clearpage

{\stsL{mart examples}
<<aabab>>=
mm = useMart("snp")
listDatasets(mm)[1:2,]
mm = useMart("snp", dataset="hsapiens_snp")
listFilters(mm)[1:3,]
args(getBM)
@
}

\clearpage

{\stsh{Summary}
\bi
\item packages exist for platform, organism, pathway/ontology annotation maps
\item SQLite database tables are fundamental entities
\item annotate package lookUp function resolves vectors of identifiers
\item AnnotationDbi get/mget; revmap facilities illustrated previously
\item SQLite annotation packages used by annaffy -- aafTableAnn -- regardless
of manufacturer
\item sqlForge tools for creating .db packages in AnnotationDbi
\ei
}


\end{document}