@@ -14,7 +14,7 @@ Version: 0.99.14

 Depends: R (>= 2.15.0), methods, GenomicRanges

 Imports: IRanges, Rsamtools (>= 1.7.4), rtracklayer (>= 1.17.15), GenomicRanges,

          GenomicFeatures, Biostrings, VariantAnnotation, tools, Biobase

-Suggests: RUnit, BSgenome.Dmelanogaster.UCSC.dm3, pasillaBamSubset, 

+Suggests: RUnit, BSgenome.Dmelanogaster.UCSC.dm3,

 	  BSgenome.Scerevisiae.UCSC.sacCer3, VariantAnnotation,

           org.Hs.eg.db, TxDb.Hsapiens.UCSC.hg19.knownGene,

           BSgenome.Hsapiens.UCSC.hg19

@@ -130,7 +130,7 @@ if (!suppressWarnings(require(BSgenome.Dmelanogaster.UCSC.dm3))) {

   library(BSgenome.Dmelanogaster.UCSC.dm3)

 }

-gmapGenomePath <- file.path(getwd(), "testGenome")

+gmapGenomePath <- file.path(getwd(), "flyGenome")

 gmapGenomeDirectory <- GmapGenomeDirectory(gmapGenomePath, create = TRUE)

 ##> gmapGenomeDirectory

 ##GmapGenomeDirectory object

@@ -151,41 +151,51 @@ gmapGenome <- GmapGenome(genome=Dmelanogaster,

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

 The \software{GSNAP} algorithm incorporates biological knowledge to

-provide accurate alignments, particularly for RNA-seq data. The

-following example synthesizes some input reads and aligns them using

-\software{GSNAP}.

-

-If you have created the D. melanogaster \Rclass{GmapGenome} package as

-detailed above and installed it, you can begin this section as

-follows:

+provide accurate alignments, particularly for RNA-seq data. In this

+section, we will align reads from an RNA-seq experiment provided in a

+fastq file to a selected region of the human genome. In this example,

+we will align to the region of the human genome containing TP53 plus

+an additional one megabase on each side of this gene.

+

+First we need to obtain the desired region of interest from the genome:

+

+<<get_TP53_coordinates>>=

+library("org.Hs.eg.db")

+library("TxDb.Hsapiens.UCSC.hg19.knownGene")

+eg <- org.Hs.eg.db::org.Hs.egSYMBOL2EG[["TP53"]]

+txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene

+tx <- transcripts(txdb, vals = list(gene_id = eg))

+roi <- range(tx) + 1e6

+@ 

-<<<load_GmapGenome, eval=FALSE>>=

-library("GmapGenome.Dmelanogaster.UCSC.dm3")

-gmapGenome <- GmapGenome.Dmelanogaster.UCSC.dm3

+Next we get the genetic sequence and use it to create a GmapGenome

+object. (Please note that the TP53_demo GmapGenome object is used by

+many examples and tests in the gmapR and VariationTools packages. If

+the object has been created before, its subsequent creation will be

+instantaneous.)

+

+<<get_TP53_seq>>=

+library("BSgenome.Hsapiens.UCSC.hg19")

+p53Seq <- getSeq(BSgenome.Hsapiens.UCSC.hg19::Hsapiens, roi,

+                 as.character = FALSE) 

+names(p53Seq) <- "TP53"

+gmapGenome <- GmapGenome(genome = p53Seq, name = "TP53_demo", create = TRUE)

 @ 

-Otherwise, you can create the \Rcode{gmapGenome} object as detailed in the

-initial section of this vignette.

+The data package \software{LungCancerLines} contains fastqs of reads

+obtained from sequencing H1993 and H2073 cell lines. We will use these

+fastqs to demonstrate \software{GSNAP} alignment with \software{gmapR}.

-<<align_with_gsnap, eval=FALSE>>=

-##synthesize paired-end FASTA input to align against the genome

-library("BSgenome.Dmelanogaster.UCSC.dm3")

-dnaStringSet <- Dmelanogaster$chr4

-makeTestFasta <- function(seqStarts, pairNum) {

-  seqEnds <- seqStarts + 74

-  nucleotides <- sapply(seq_along(seqStarts),

-                        function(i) as.character(subseq(dnaStringSet, seqStarts[i], seqEnds[i]))

-                        )

-  fastaHeaders <- paste0(">", "fakeRead", seq_along(nucleotides), "#0/", pairNum)

-  fastaLines <- paste(fastaHeaders, nucleotides, sep="\n")

-  tmpFasta <- paste0(tempfile(), ".", pairNum, ".fa")

-  writeLines(fastaLines, con=tmpFasta)

-  tmpFasta

-}

-seqStarts <- seq(1, 100000, by = 10000) + length(dnaStringSet) * 2 / 3

-tmpFasta1 <- makeTestFasta(seqStarts, '1')

-tmpFasta2 <- makeTestFasta(seqStarts + 125, '2')

+<<get_lung_cancer_fastqs>>=

+library("LungCancerLines")

+fastqs <- LungCancerFastqFiles()

+@ 

+\software{GSNAP} is highly configurable. Users create

+\Rclass{GsnapParam} objects to specify desired \software{GSNAP}

+behavior. 

+

+<<create_gsnapParam>>=

 ##specify how GSNAP should behave using a GsnapParam object

 gsnapParam <- GsnapParam(genome = gmapGenome,

                          unique_only = FALSE,

@@ -194,10 +204,15 @@ gsnapParam <- GsnapParam(genome = gmapGenome,

                          npaths = 10,

                          novelsplicing = FALSE, splicing = NULL, 

                          nthreads = 1,

-                         batch = 2L)

+                         batch = 2L,

+                         gunzip=TRUE)

+@ 

-gsnapOutput <- gsnap(input_a=tmpFasta1,

-                     input_b=tmpFasta2,

+Now we are ready to align.

+

+<<align_with_gsnap, eval=FALSE>>=

+gsnapOutput <- gsnap(input_a=fastqs["H1993.first"],

+                     input_b=fastqs["H1993.last"],

                      params=gsnapParam)

 ##gsnapOutput

 ##An object of class "GsnapOutput"

@@ -244,29 +259,41 @@ alignments that \software{GSNAP} outputs along with other utilities.

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

 Running the \Rmethod{bam\_tally} method will return a \Rclass{GRanges}

-of information per nucleotide. Below is a basic example. See the

-documentation for the \Rmethod{bam\_tally} method for details.

+of information per nucleotide. Below is an example demonstrating how

+to find variants in the TP53 gene of the human genome. See the

+documentation for the \Rmethod{bam\_tally} method for more details.

+

+\software{gmapR} provides access to a demo genome for examples. This

+genome encompasses the TP53 gene along with a 1-megabase flanking

+region on each side.

+<<TP53Genome_accessor>>=

+genome <- TP53Genome()

+@ 

+

+The \software{LungCancerLines} R package contains a BAM file of reads

+aligned to the TP53 region of the human genome. We'll use this file to

+demonstrate the use of \software{bam_tally} through the

+\software{gmapR} package. The resulting data structure will contain

+the needed information such as number of alternative alleles, quality

+scores, and nucleotide cycle for each allele.

 <<run_bamtally, eval=FALSE>>=

 library(gmapR)

-library(GmapGenome.Hsapiens.UCSC.hg19)

-genome <- GmapGenome.Hsapiens.UCSC.hg19

-bam_file <- file.path(system.file(package = "gmapR", mustWork=TRUE),

-                      "extdata/bam_tally_test.bam")

+bam_file <- system.file("extdata/H1993.analyzed.bam", 

+                        package="LungCancerLines", mustWork=TRUE)

 breaks <- c(0L, 15L, 60L, 75L)

 bqual <- 56L

 mapq <- 13L

 param <- BamTallyParam(genome, cycle_breaks = breaks,

-                       high_quality_cutoff = bqual,

+                       high_base_quality = bqual,

                        minimum_mapq = mapq,

                        concordant_only = FALSE, unique_only = FALSE,

                        primary_only = FALSE,

                        min_depth = 0L, variant_strand = 1L,

                        ignore_query_Ns = TRUE,

                        indels = FALSE)

-

 tallies <-bam_tally(bam_file,

                     param)

 @ 

@@ -330,26 +357,28 @@ repository, etc. After installation,

 \Rcode{library("GmapGenome.Hsapiens.UCSC.hg19")} will load a

 \Rclass{GmapGenome} object that has the same name as the package.

-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

-\section{Aligning with SNP Tolerance}

-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

-Both \software{GSNAP} and \software{GMAP} have the ability to perform

-alignment without penalizing against a set of known SNPs. In this

-section, we work through an example of aligning with this SNP

-tolerance. By using SNP-tolerance, we'll perform alignments without

-penalizing reads that match SNPs contained in the provided SNPs. In

-this case, we'll use dbSNP to provide the SNPs.

-This example will use hg19. If you have not installed

-\Rcode{GmapGenome.Hsapiens.UCSC.hg19}, please refer to the section

-\ref{create_hg19_GmapGenome}.

-

-First we load the Hsapiens \Rclass{GmapGenome} object:

+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

+%\section{Aligning with SNP Tolerance}

+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

-<<<load_GmapGenomeHsapiens>>=

-library("GmapGenome.Hsapiens.UCSC.hg19")

-@ 

+%Both \software{GSNAP} and \software{GMAP} have the ability to perform

+%alignment without penalizing against a set of known SNPs. In this

+%section, we work through an example of aligning with this SNP

+%tolerance. By using SNP-tolerance, we'll perform alignments without

+%penalizing reads that match SNPs contained in the provided SNPs. In

+%this case, we'll use dbSNP to provide the SNPs.

+%

+%This example will use hg19. If you have not installed

+%\Rcode{GmapGenome.Hsapiens.UCSC.hg19}, please refer to the section

+%\ref{create_hg19_GmapGenome}.

+%

+%First we load the Hsapiens \Rclass{GmapGenome} object:

+%

+%<<<load_GmapGenomeHsapiens>>=

+%library("GmapGenome.Hsapiens.UCSC.hg19")

+%@ 

 <<SessionInfo>>=

 sessionInfo()