The last step of a classical Roar analyses: it returns a dataframe containing m/M values, roar values, pvalues and estimates of expression (a measure recalling FPKM).
The resulting dataframe will be identical to that returned by
with two columns added: "treatmentValue" and "controlValue". These columns will contain a number
that indicates the level of expression of the relative gene in the treatment (or control) condition.
RoarDataset this number derives from the counts (averages across samples when
applicable) obtained for the PRE portion of the gene and is similar
to the RPKM standard measure of expression used in RNAseq experiment. Specifically
we correct the counts on the PRE portions dividing them by portion length and total numer
of reads aligned on all PRE portions and the multiply the results for 1000000000.
See the vignette for more details.
RoarDatasetMultipleAPA the same procedure is applied to all the possible
PRE choices for genes. Note that summing all the counts for every PRE portion assigned to a gene
could lead to count some reads multiple times when summing all the PRE portions counts therefore
this measure is not completely comparable with the one obtained with the single APA analysis.
The length column added in this case contains the length of the PRE portions (counting only
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
library(GenomicAlignments) gene_id <- c("A_PRE", "A_POST", "B_PRE", "B_POST") features <- GRanges( seqnames = Rle(c("chr1", "chr1", "chr2", "chr2")), strand = strand(rep("+", length(gene_id))), ranges = IRanges( start=c(1000, 2000, 3000, 3600), width=c(1000, 900, 600, 300)), DataFrame(gene_id) ) rd1 <- GAlignments("a", seqnames = Rle("chr1"), pos = as.integer(1000), cigar = "300M", strand = strand("+")) rd2 <- GAlignments("a", seqnames = Rle("chr1"), pos = as.integer(2000), cigar = "300M", strand = strand("+")) rd3 <- GAlignments("a", seqnames = Rle("chr2"), pos = as.integer(3000), cigar = "300M", strand = strand("+")) rds <- RoarDataset(list(c(rd1,rd2)), list(rd3), features) rds <- countPrePost(rds, FALSE) rds <- computeRoars(rds) rds <- computePvals(rds) dat <- fpkmResults(rds)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.