CAGEr (version 1.14.0)

getCTSS: Reading CAGE data from input file(s) and detecting TSSs

Description

Reads input CAGE datasets into CAGEset object, constructs CAGE transcriptions start sites (CTSSs) and counts number of CAGE tags supporting every CTSS in each input experiment. Preprocessing and quality filtering of input CAGE tags, as well as correction of CAGE-specific 'G' nucleotide addition bias can be also performed before constructing TSSs.

Usage

getCTSS(object, sequencingQualityThreshold = 10,  mappingQualityThreshold = 20, removeFirstG = TRUE,  correctSystematicG = TRUE)

Arguments

object

A CAGEset object

sequencingQualityThreshold, mappingQualityThreshold

Only CAGE tags with average sequencing quality >= sequencingQualityThreshold and mapping quality >= mappingQualityThreshold are kept. Used only if inputFileType(object) == "bam" or inputFileType(object) == "bamPairedEnd", i.e when input files are BAM files of aligned sequenced CAGE tags, otherwise ignored. If there are no sequencing quality values in the BAM file (e.g. HeliScope single molecule sequencer does not return sequencing qualities) all reads will by default have this value set to -1. Since the default value of sequencingQualityThreshold is 10, all the reads will consequently be discarded. To avoid this behaviour and keep all sequenced reads set sequencingQualityThreshold to -1 when processing data without sequencing qualities. If there is no information on mapping quality in the BAM file (e.g. software used to align CAGE tags to the referent genome does not provide mapping quality) the mappingQualityThreshold parameter is ignored. In case of paired-end sequencing BAM file (i.e. inputFileType(object) == "bamPairedEnd") only the first mate of the properly paired reads (i.e. the five prime end read) will be read and subject to specified thresholds.

removeFirstG

Logical, should the first nucleotide of the CAGE tag be removed in case it is a G and it does not map to the referent genome (i.e. it is a mismatch). Used only if inputFileType(object) == "bam" or inputFileType(object) == "bamPairedEnd", i.e when input files are BAM files of aligned sequenced CAGE tags, otherwise ignored. See Details.

correctSystematicG

Logical, should the systematic correction of the first G nucleotide be performed for the positions where there is a G in the CAGE tag and G in the genome. This step is performed in addition to removing the first G of the CAGE tags when it is a mismatch, i.e. this option can only be used when removeFirstG = TRUE, otherwise it is ignored. The frequency of adding a G to CAGE tags is estimated from mismatch cases and used to systematically correct the G addition for positions with G in the genome. Used only if inputFileType(object) == "bam" or inputFileType(object) == "bamPairedEnd", i.e when input files are BAM files of aligned sequenced CAGE tags, otherwise ignored. See Details.

Value

The slots librarySizes, CTSScoordinates and tagCountMatrix of the provided CAGEset object will be occupied by the information on CTSSs created from input CAGE files.

Details

In the CAGE experimental protocol an additional G nucleotide is often attached to the 5' end of the tag by the template-free activity of the reverse transcriptase used to prepare cDNA (Harbers and Carninci, Nature Methods 2005). In cases where there is a G at the 5' end of the CAGE tag that does not map to the corresponding genome sequence, it can confidently be considered spurious and should be removed from the tag to avoid misannotating actual TSS. Thus, setting removeFirstG = TRUE is highly recommended.

However, when there is a G both at the beginning of the CAGE tag and in the genome, it is not clear whether the original CAGE tag really starts at this position or the G nucleotide was added later in the experimental protocol. To systematically correct CAGE tags mapping at such positions, a general frequency of adding a G to CAGE tags can be calculated from mismatch cases and applied to estimate the number of CAGE tags that have G added and should actually start at the next nucleotide/position. The option correctSystematicG is an implementation of the correction algorithm described in Carninci et al., Nature Genetics 2006, Supplementary Information section 3-e.

References

Harbers and Carninci (2005) Tag-based approaches for transcriptome research and genome annotation, Nature Methods 2(7):495-502.

Carninci et al. (2006) Genome-wide analysis of mammalian promoter architecture and evolution, Nature Genetics 38(7):626-635.

See Also

CTSScoordinates

CTSStagCount

Examples

library(BSgenome.Drerio.UCSC.danRer7)

pathsToInputFiles <- system.file("extdata", c("Zf.unfertilized.egg.chr17.ctss",
"Zf.30p.dome.chr17.ctss", "Zf.prim6.rep1.chr17.ctss"), package="CAGEr")
labels <- paste("sample", seq(1,3,1), sep = "")

myCAGEset <- new("CAGEset", genomeName = "BSgenome.Drerio.UCSC.danRer7",
inputFiles = pathsToInputFiles, inputFilesType = "ctss", sampleLabels = labels)

getCTSS(myCAGEset)