## ----style, echo = FALSE, results = 'asis'-------------------------------
BiocStyle::markdown()

## ----, echo = FALSE------------------------------------------------------
apiKey <- Sys.getenv("GOOGLE_API_KEY")
if (nchar(apiKey) == 0) {
  warning(paste("To build this vignette, please setup the environment variable",
                "GOOGLE_API_KEY with the public API key from your Google",
                "Developer Console before loading the GoogleGenomics package,",
                "or run GoogleGenomics::authenticate."))
  knitr::knit_exit()
}

## ------------------------------------------------------------------------
suppressPackageStartupMessages(library(VariantAnnotation))
fl <- system.file("extdata", "chr22.vcf.gz", package="VariantAnnotation")
vcf <- readVcf(fl, "hg19")
vcf <- renameSeqlevels(vcf, c("22"="chr22"))
vcf

## ------------------------------------------------------------------------
# Authenticated on package load from the env variable GOOGLE_API_KEY.
suppressPackageStartupMessages(library(GoogleGenomics))

# TODO Right now we're just getting a few variants.  Later update this to retrieve them all.
system.time({
granges <- getVariants(datasetId="10473108253681171589",
                       chromosome="22",
                       start=50300077,
                       end=50303000,         # TODO end=50999964
                       converter=variantsToGRanges)
})

## ------------------------------------------------------------------------
vcf <- vcf[1:length(granges)] # Truncate the VCF data

suppressPackageStartupMessages(library(testthat))
expect_equal(start(granges), start(vcf))
expect_equal(end(granges), end(vcf))
expect_equal(as.character(granges$REF), as.character(ref(vcf)))
expect_equal(as.character(unlist(granges$ALT)), as.character(unlist(alt(vcf))))
expect_equal(granges$QUAL, qual(vcf))
expect_equal(granges$FILTER, filt(vcf))

## ------------------------------------------------------------------------
suppressPackageStartupMessages(library(TxDb.Hsapiens.UCSC.hg19.knownGene))
txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene


rd <- rowData(vcf)
vcf_locations <- locateVariants(rd, txdb, CodingVariants())
vcf_locations

granges_locations <- locateVariants(granges, txdb, CodingVariants())
granges_locations

expect_equal(granges_locations, vcf_locations)

## ------------------------------------------------------------------------
suppressPackageStartupMessages(library(BSgenome.Hsapiens.UCSC.hg19))
vcf_coding <- predictCoding(vcf, txdb, seqSource=Hsapiens)
vcf_coding

granges_coding <- predictCoding(rep(granges, elementLengths(granges$ALT)),
                                txdb,
                                seqSource=Hsapiens,
                                varAllele=unlist(granges$ALT, use.names=FALSE))

granges_coding

expect_equal(as.matrix(granges_coding$REFCODON), as.matrix(vcf_coding$REFCODON))
expect_equal(as.matrix(granges_coding$VARCODON), as.matrix(vcf_coding$VARCODON))
expect_equal(granges_coding$GENEID, vcf_coding$GENEID)
expect_equal(granges_coding$CONSEQUENCE, vcf_coding$CONSEQUENCE)


## ------------------------------------------------------------------------
suppressPackageStartupMessages(library(org.Hs.eg.db))
annots <- select(org.Hs.eg.db,
                 keys=granges_coding$GENEID,
                 keytype="ENTREZID",
                 columns=c("SYMBOL", "GENENAME","ENSEMBL"))
cbind(elementMetadata(granges_coding), annots)

## ------------------------------------------------------------------------
sessionInfo()