RSEMCalculateExpression: Use the RSEM tool to annotate and quantify the reads

In RGLab/RNASeqPipelineR: Construct Standardized pipelines for processing RNASeq data

Description

Use the RSEM tool to annotate and quantify the reads

Usage

RSEMCalculateExpression(parallel_threads = 6, bowtie_threads = 1,
  paired = FALSE, frag_mean = NULL, frag_sd = NULL, nchunks = 10,
  days_requested = 5, slurm = FALSE, slurm_partition = NULL,
  ram_per_node = bowtie_threads * parallel_threads * 1200,
  fromBAM = FALSE, fromSTAR = FALSE, mail = NULL, force = FALSE)

Arguments

parallel_threads	integer specify how many parallel processes to spawn
bowtie_threads	integer specify how many threads bowtie should use.
paired	logical specify whether you have paried reads or not.
frag_mean	numeric
frag_sd	numeric For single ended reads, specifying these might make calculations of effect length more effective. Optional.
nchunks	integer number of chunks to split the files for a slurm job. Ignored if slurm = FALSE
days_requested	integer number of days requested for the job (when submitting a slurm job). Ignored if slurm = FALSE
slurm	logical if TRUE job is submitted as a slurm batch job, otherwise it's run on the local machine. Slurm jobs will honour the nchunks and days_requested arguments.
slurm_partition	character the slurm partition to submit to. Ignored if slurm=FALSE
ram_per_node	numeric The number of Mb per node. Ignored if slurm=FALSE. Default of parallel_threads*bowtie_threads*1000
fromBAM	logical if TRUE then RSEM will attempt to use previously aligned BAM files, in the BAM directory, rather than fastq files. The file names expected to end with .transcript.bam. See RSEM documentation for the format these files must obey.
fromSTAR	logical if TRUE then RSEM will use the STAR notation for the BAM files
mail	email address to send failure message to, if desired.
force	If TRUE then quantify all FASTQ/BAM files else only quantify those files that have not been quantified.

Details

Uses RSEM to quantify reads in FASTQ files against the reference genome. Optionally you can specify paired end reads. The code assumes paired reads have fastq files that differ by one character (i.e. sampleA_read1.fastq, sampleA_read2.fastq) and will perform matching of paired fastq files based on that assumption using string edit distance. Read 1 is assumed to be upstream and read 2 is assumed to be downstream. The number of parallel_threads*bowtie_threads should not be more than the number of cores available on your system.

Note

The amount of memory requested should be set to bowtie_threads*parallel_threads*1G as this is the default requested by samtools for sorting. If insufficient memory is requested, the bam files will not be created successfully.

RGLab/RNASeqPipelineR documentation built on Jan. 19, 2020, 12:31 a.m.