diff metaseqr_biotypes_run346_unfiltered.txt metaseqr_biotypes_run346.txt
no diff

Dear Pantelis,

I've analyzed one RNAseq and 2 3UTRseq runs of the ITlab
wrt percentage of reads in all biotypes.
The rRNA fraction in neglectable in all cases:

RNAseq_run346  0.002%
3utrSeq_run329 0.02 %
3utrSeq_run391 0 %

All biotype percentages are attached.

BW,
Martin
library(metaseqR)
metaseqr(
    sample.list=file.path(the.path,"targets.txt"),
    contrast=c("Set9ko12_vs_Set9ko34"),
    annotation="download",
    org="mm10",
    count.type="gene",
    normalization="deseq",
    statistics="deseq",
    fig.format=c("png","pdf"),
    export.where=file.path(the.path,"metaseqr_rRNA"),
    restrict.cores=0.5,
    qc.plots=c(
        "mds","biodetection","countsbio","saturation","readnoise","filtered",
        "correl","pairwise","boxplot","gcbias","lengthbias","meandiff",
        "meanvar","biodist","volcano","deheatmap"
    ),
    exon.filters=NULL,
    gene.filters=list(
       length=list(
            length=500
        ),
        avg.reads=list(
            average.per.bp=100,
            quantile=0.25
        ),
        expression=list(
            median=TRUE,
            mean=FALSE,
            quantile=NA,
            known=NA,
            custom=NA
        ),
        biotype=get.defaults("biotype.filter","mm10")
    ),
    pcut=0.05,
    export.what=c("annotation","p.value","adj.p.value","fold.change",
		"counts","flags"),
    export.scale=c("log2","rpgm"),
    export.values="normalized",
    export.counts.table=TRUE,
	report.top=0.05
)

the.path <- "/data/images/proton2/run346/www"
 zcat /data/images/proton2/run346/www/metaseqr_rRNA/lists/raw_counts_table.txt.gz | awk -f /data/images/proton2/get-biotype-pct1.awk | sort -k2,2nr > metaseqr_biotypes_run346.txt
total reads124600704


zcat /data/images/proton2/run329/www/metaseqr_quantseq_wt_vs_ko/lists/raw_counts_table.txt.gz | awk -f /data/images/proton2/get-biotype-pct1.awk | sort -k2,2nr > metaseqr_rRNA_biotypes.txt >  metaseqr_rRNA_biotypes_3utrSeq_run329.txt

# Read transcript data from external file
transcript.data <- read.delim("/data/results/tools/rnaseq/metaseqr/mm10/transcript_data_mm10.txt.gz")
rownames(transcript.data) <- as.character(transcript.data$transcript_id)

# metaseqR related variables outside the pipeline
multic <- check.parallel(0.5) # If wish to use multiple cores
assign("VERBOSE",TRUE,envir=metaseqR:::meta.env)

# Read targets files
message("Reading targets file...")
targets <- read.targets("targets.txt")
# Do the counting based with bam files in the targets file
r2c.out <- read2count(targets,transcript.data,has.all.fields=TRUE,multic=multic)

# Create a counts table to be passed as "embedded" annotation
the.counts <- cbind(r2c.out$mergedann,r2c.out$counts)

    contrast=c("Set9ko12_vs_Set9ko34"),


metaseqr(
	counts=the.counts,
    sample.list=targets$samples,
    contrast=c("Set9ko12_vs_Set9ko34"),
    annotation="embedded",
	gene.file="/data/results/tools/rnaseq/metaseqr/mm10/gene_data_mm10.txt.gz",
	id.col=4, # required with embedded annotation
	gc.col=5, # required with embedded annotation
	bt.col=8, # required with embedded annotation
	name.col=7, # required with embedded annotation
    org="custom",
    count.type="gene",
    normalization="deseq", # or whatever supported
    statistics="deseq", # or whatever supported, more than one also
    fig.format=c("png","pdf"),
    export.where=file.path(the.path,"metaseqr_rRNA"),
    restrict.cores=0.5, # fraction of available cores to use
    qc.plots=c(
        "mds","biodetection","countsbio","saturation","readnoise","filtered",
        "correl","pairwise","boxplot","gcbias","lengthbias","meandiff",
        "meanvar","biodist","volcano","deheatmap"
    ),
    exon.filters=NULL,
    gene.filters=list(
       length=list(
            length=500
        ),
        avg.reads=list(
            average.per.bp=100,
            quantile=0.25
        ),
        expression=list(
            median=TRUE,
            mean=FALSE,
            quantile=NA,
            known=NA,
            custom=NA
        ),  # it's the default anyway
        biotype=get.defaults("biotype.filter","mm10")
    ),
    pcut=0.05, # only for the truncated significant list output, all results are exported anyway
    export.what=c("annotation","p.value","adj.p.value","fold.change","stats",
        "counts","flags"), # if you use pandora, the fields "meta.p.value" and "adj.meta.p.value" should be added
    export.scale=c("log2","rpgm"),
    export.values="normalized",
    export.stats="mean"
)



#


metaseqr(
    sample.list=file.path(the.path,"targets.txt"),
    contrast=c("Set9ko12_vs_Set9ko34"),
    annotation="download",
    org="mm10",
    count.type="gene",
    normalization="deseq",
    statistics="deseq",
    fig.format=c("png","pdf"),
    export.where=file.path(the.path,"metaseqr_rRNA"),
    restrict.cores=0.5,
    qc.plots=c(
        "mds","biodetection","countsbio","saturation","readnoise","filtered",
        "correl","pairwise","boxplot","gcbias","lengthbias","meandiff",
        "meanvar","biodist","volcano","deheatmap"
    ),
    exon.filters=NULL,
    gene.filters=list(
       length=list(
            length=500
        ),
        avg.reads=list(
            average.per.bp=100,
            quantile=0.25
        ),
        expression=list(
            median=TRUE,
            mean=FALSE,
            quantile=NA,
            known=NA,
            custom=NA
        ),
        biotype=get.defaults("biotype.filter","mm10")
    ),
    pcut=0.05,
    export.what=c("annotation","p.value","adj.p.value","fold.change",
		"counts","flags"),
    export.scale=c("log2","rpgm"),
    export.values="normalized",
    export.counts.table=TRUE,
	report.top=0.05
)