buildAllAnnotationFiles: Create All Annotation-dependant Files for a DuffyNGS Target...

In robertdouglasmorrison/DuffyNGS: Duffy Lab NGS Analysis Pipeline for RNA-seq, DNA-seq, ChIP-seq, and RIP-seq

Create All Annotation-dependant Files for a DuffyNGS Target Organism(s)

Description

Build all needed files for running the various Bowtie pipeline pieces for a target organism(s). This will construct genomic Bowtie indexes, RiboClearing maps and indexes, Splice Junction maps and indexs, and cross species detectability objects.

Usage

buildAllAnnotationFiles(speciesIDSet, genomicFastaFileSet, 
	fastaPatternSet = "\.fasta$", optionsFileSet = "Options.txt", 
	outPath = ".", steps = 1:4, riboTailSize = 36, 
	exonFragmentSize = 36, maxExonSkipSet = 2, intronSplicesTooSet = FALSE, 
	detectableReadSize = 32, chunkSize = 1e+06, 
	otherSpeciesIDSet = "Hs_grc", otherGenomicIndexSet = "Hs.genomic_idx", 
	verbose = TRUE, debug = FALSE)

Arguments

speciesIDSet	a character vector of the SpeciesIDs that comprise one new target
genomicFastaFileSet	a character vector (same length as speciesIDset) of the FASTA files or folders that are the genomic definition of each species.
fastaPatternSet	a character vector (same length again) of regular expressions for finding FASTA files in the folder, for species that have each chromosome in separate files.
optionsFileSet	a character vector of OptionsFile names. If more than one species, this vector must have an options file for each, plus one extra for the combined target.
outPath	destination folder for local copies of all created files
steps	See details. A vector in integers in the range 1:4, to specify which componment files to build. Since each step can be quite time consuming (or error prone), they can be done one at a time or just as new annotation files (MapSets) dictate.
riboTailSize	the number of bases to extend past the ends of genes being targeted for RiboClearing. This allows reads that overlap the gene ends to still align, and thus be cleared. The value should be about 75% of a typical read length.
exonFragmentSize	the number of bases into each exon to use for building the Splice Junction index. The value should be about 90% of a typical read length, so any aligned reads must be spanning 2 exons.
maxExonSkipSet	the maximum number of exons to be skipped, for the alternative splicing exon pairs.
intronSplicesTooSet	logical, build alternative splice junctions using both introns and exons.
detectableReadSize	number of bases in a canonical read, for measuring in-species uniqueness and cross-species detectability.
chunkSize	buffer size, for the detectability step
otherSpeciesIDSet	a character vector of 'second' SpeciesIDs, for building cross-detectability files
otherGenomicIndexSet	a character vector (same length as otherSpeciesIDset) of the Bowtie genomic indexes for each 'second' species.
verbose	logical, print lots of messages during progress
debug	logical, perform sanity checks during processing

Details

This is an umbrella function that attempts to build all needed files for a new target organism(s) and/or new read length data sets. It builds all needed Bowtie indexes (Genomic, RiboClearing, Splice Junctions), all needed alignment decoding maps (RiboClearing, Splice Junctions), and all 2-species detectability information.

The key focus is to easily combine multiple organisms into one joint target index family in a way that seemlessly creates all needed files for future analysis of these organisms.

For the non-genomic indexes, the typical raw read lengths impacts the settings for building these indexes, to optimize the align-ability of the reads. For this reason, it is often wise to make separate indexes if you will work with very different read lengths (i.e. separate riboClear and splice junction indexes for short (~40bp) and long (~100bp) reads.)

Step=1: Build Genomic indexes for each organism, and a joint genomic index if speciesIDSet is longer than one.

Step=2: Build Ribo Clearing indexes and maps for each organism, and a joint RiboClearing index if speciesIDSet is longer than one.

Step=3: Build Splice Junction indexes and maps for each organism, and a joint SpliceJunction index if speciesIDSet is longer than one.

Step=4: Build detectability files.

Value

A family of files written to disk, that may need to be moved to the appropriate folders for use in subsequent pipeline runs.

Note

Indexes and maps need to be moved to the folder of Bowtie indexes, and detectability files need to be moved to the /data subfolder of the the DuffyNGS package installation location.

Author(s)

Bob Morrison

robertdouglasmorrison/DuffyNGS documentation built on March 24, 2024, 4:16 p.m.