Package 'ORFik'

Description

Main goals:

1. Finding Open Reading Frames (very fast) in the genome of interest or on the set of transcripts/sequences.

2. Utilities for metaplots of RiboSeq coverage over gene START and STOP codons allowing to spot the shift.

3. Shifting functions for the RiboSeq data.

4. Finding new Transcription Start Sites with the use of CageSeq data.

5. Various measurements of gene identity e.g. FLOSS, coverage, ORFscore, entropy that are recreated based on many scientific publications.

6. Utility functions to extend GenomicRanges for faster grouping, splitting, tiling etc.

Author(s)

Maintainer: Haakon Tjeldnes [email protected] [data contributor]

Authors:

• Kornel Labun [email protected] [copyright holder]

Other contributors:

• Michal Swirski [email protected] [contributor]

• Katarzyna Chyzynska [email protected] [contributor, data contributor]

• Yamila Torres Cleuren [email protected] [contributor, thesis advisor]

• Eivind Valen [email protected] [thesis advisor, funder]

See Also

Useful links:

• https://github.com/Roleren/ORFik

• Report bugs at https://github.com/Roleren/ORFik/issues

Description

Usefull to see if short ORFs prediction is dependent on length.
Split cds first in two, a start part and stop part. Then say how large the two parts can be and merge them together. It will sample a value in range give.
Parts will be forced to not overlap and can not extend outside original cds

Usage

artificial.orfs(
  cds,
  start5 = 1,
  end5 = 4,
  start3 = -4,
  end3 = 0,
  bin.if.few = TRUE
)

Arguments

Details

If artificial cds length is not divisible by 2, like 3 codons, the second codon will always be from the start region etc.
Also If there are many very short original cds, the distribution will be skewed towards more smaller artificial cds.

Value

GRangesList of new ORFs (sorted: + strand increasing start, - strand decreasing start)

Examples

txdb <- ORFik.template.experiment()
#cds <- loadRegion(txdb, "cds")
## To get enough CDSs, just replicate them
# cds <- rep(cds, 100)
#artificial.orfs(cds)

Description

For all cds in txdb, that does not have a 5' leader: Start at 1 base upstream of cds and use CAGE, to assign leader start. All these leaders will be 1 exon based, if you really want exon splicings, you can use exon prediction tools, or run sequencing experiments.

Usage

assignTSSByCage(
  txdb,
  cage,
  extension = 1000,
  filterValue = 1,
  restrictUpstreamToTx = FALSE,
  removeUnused = FALSE,
  preCleanup = TRUE,
  pseudoLength = 1
)

Arguments

Details

Given a TxDb object, reassign the start site per transcript using max peaks from CageSeq data. A max peak is defined as new TSS if it is within boundary of 5' leader range, specified by 'extension' in bp. A max peak must also be higher than minimum CageSeq peak cutoff specified in 'filterValue'. The new TSS will then be the positioned where the cage read (with highest read count in the interval). If no CAGE supports a leader, the width will be set to 1 base.

Value

a TxDb obect of reassigned transcripts

See Also

Other CAGE: reassignTSSbyCage(), reassignTxDbByCage()

Examples

txdbFile <- system.file("extdata", "hg19_knownGene_sample.sqlite",
 package = "GenomicFeatures")
cagePath <- system.file("extdata", "cage-seq-heart.bed.bgz",
 package = "ORFik")

## Not run: 
  assignTSSByCage(txdbFile, cagePath)
  #Minimum 20 cage tags for new TSS
  assignTSSByCage(txdbFile, cagePath, filterValue = 20)
  # Create pseudo leaders for the ones without hits
  assignTSSByCage(txdbFile, cagePath, pseudoLength = 100)
  # Create only pseudo leaders (in example 2 leaders are added)
  assignTSSByCage(txdbFile, cage = NULL, pseudoLength = 100)

## End(Not run)

Description

Map range coordinates between features in the genome and transcriptome (reference) space.

Usage

asTX(
  grl,
  reference,
  ignore.strand = FALSE,
  x.is.sorted = TRUE,
  tx.is.sorted = TRUE
)

Arguments

Details

Similar to GenomicFeatures' pmapToTranscripts, but in this version the grl ranges are compared to reference ranges with same name, not by index. And it has a security fix.

Value

a GRangesList in transcript coordinates

See Also

Other ExtendGenomicRanges: coveragePerTiling(), extendLeaders(), extendTrailers(), reduceKeepAttr(), tile1(), txSeqsFromFa(), windowPerGroup()

Examples

seqname <- c("tx1", "tx2", "tx3")
seqs <- c("ATGGGTATTTATA", "AAAAA", "ATGGGTAATA")
grIn1 <- GRanges(seqnames = "1",
                 ranges = IRanges(start = c(21, 10), end = c(23, 19)),
                 strand = "-")
grIn2 <- GRanges(seqnames = "1",
                 ranges = IRanges(start = c(1), end = c(5)),
                 strand = "-")
grIn3 <- GRanges(seqnames = "1",
                 ranges = IRanges(start = c(1010), end = c(1019)),
                 strand = "-")
grl <- GRangesList(grIn1, grIn2, grIn3)
names(grl) <- seqname
# Find ORFs
test_ranges <- findMapORFs(grl, seqs,
                 "ATG|TGG|GGG",
                 "TAA|AAT|ATA",
                 longestORF = FALSE,
                 minimumLength = 0)
# Genomic coordinates ORFs
test_ranges
# Transcript coordinate ORFs
asTX(test_ranges, reference = grl)
# seqnames will here be index of transcript it came from

Get library variable names from ORFik experiment

Description

What will each sample be called given the columns of the experiment? A column is included if more than 1 unique element value exist in that column.

Usage

bamVarName(
  df,
  skip.replicate = length(unique(df$rep)) == 1,
  skip.condition = length(unique(df$condition)) == 1,
  skip.stage = length(unique(df$stage)) == 1,
  skip.fraction = length(unique(df$fraction)) == 1,
  skip.experiment = !df@expInVarName,
  skip.libtype = FALSE,
  fraction_prepend_f = TRUE
)

Arguments

Value

variable names of libraries (character vector)

See Also

Other ORFik_experiment: ORFik.template.experiment(), ORFik.template.experiment.zf(), create.experiment(), experiment-class, filepath(), libraryTypes(), organism,experiment-method, outputLibs(), read.experiment(), save.experiment(), validateExperiments()

Examples

df <- ORFik.template.experiment()
bamVarName(df)

## without libtype
bamVarName(df, skip.libtype = TRUE)
## Without experiment name
bamVarName(df, skip.experiment = TRUE)

Description

Open SRA in browser for specific bioproject

Usage

browseSRA(x, browser = getOption("browser"))

Arguments

Value

invisible(NULL), opens webpage only

See Also

Other sra: download.SRA(), download.SRA.metadata(), download.ebi(), get_bioproject_candidates(), install.sratoolkit(), rename.SRA.files()

Examples

#browseSRA("PRJNA336542")

#' # For windows make sure a valid browser is defined:
browser <- getOption("browser")
#browseSRA("PRJNA336542", browser)

Description

Per AA / codon, analyse the coverage, get a multitude of features. For both A sites and P-sites (Input reads must be P-sites for now) This function takes inspiration from the codonDT paper, and among others returns the negative binomial estimates, but in addition many other features.

Usage

codon_usage(
  reads,
  cds,
  mrna,
  faFile,
  filter_table,
  filter_cds_mod3 = TRUE,
  min_counts_cds_filter = max(min(quantile(filter_table, 0.5), 1000), 1000),
  with_A_sites = TRUE,
  aligned_position = "center",
  code = GENETIC_CODE
)

Arguments

Details

The primary column to use is "mean_txNorm", this is the fair normalized score.

Value

a data.table of rows per codon / AA. All values are given per library, per site (A or P), sorted by the mean_txNorm_percentage column of the first library in the set, the columns are:

• variable (character)Library name

• seq (character)Amino acid:codon

• sum (integer)total counts per seq

• sum_txNorm (integer)total counts per seq normalized per tx

• var (numeric)variance of total counts per seq

• N (integer)total number of codons of that type

• mean_txNorm (numeric)Default use output, the fair codon usage, normalized both for gene and genome level for codon and read counts

• ...

• alpha (numeric)dirichlet alpha MOM estimator (imagine mean and variance of probability in 1 value, the lower the value, the higher the variance, mean is decided by the relative value between samples)

• sum_txNorm (integer)total counts per seq normalized per tx

• relative_to_max_score (integer)Percentage use of codon

• type (factor(character))Either "P" or "A"

References

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7196831/

See Also

Other codon: codon_usage_exp(), codon_usage_plot()

Examples

df <- ORFik.template.experiment()[9:10,] # Subset to 2 Ribo-seq libs

## For single library
reads <- fimport(filepath(df[1,], "pshifted"))
cds <- loadRegion(df, "cds", filterTranscripts(df))
mrna <- loadRegion(df, "mrna", names(cds))
filter_table <- assay(countTable(df, type = "summarized")[names(cds)])
faFile <- findFa(df)
res <- codon_usage(reads, cds, mrna, faFile = faFile,
             filter_table = filter_table, min_counts_cds_filter = 10)

Description

Per AA / codon, analyse the coverage, get a multitude of features. For both A sites and P-sites (Input reads must be P-sites for now) This function takes inspiration from the codonDT paper, and among others returns the negative binomial estimates, but in addition many other features.

Usage

codon_usage_exp(
  df,
  reads,
  cds = loadRegion(df, "cds", filterTranscripts(df)),
  mrna = loadRegion(df, "mrna", names(cds)),
  filter_cds_mod3 = TRUE,
  filter_table = assay(countTable(df, type = "summarized")[names(cds)]),
  faFile = df@fafile,
  min_counts_cds_filter = max(min(quantile(filter_table, 0.5), 1000), 1000),
  with_A_sites = TRUE,
  code = GENETIC_CODE,
  aligned_position = "center"
)

Arguments

Details

The primary column to use is "mean_txNorm", this is the fair normalized score.

Value

a data.table of rows per codon / AA. All values are given per library, per site (A or P), sorted by the mean_txNorm_percentage column of the first library in the set, the columns are:

• variable (character)Library name

• seq (character)Amino acid:codon

• sum (integer)total counts per seq

• sum_txNorm (integer)total counts per seq normalized per tx

• var (numeric)variance of total counts per seq

• N (integer)total number of codons of that type

• mean_txNorm (numeric)Default use output, the fair codon usage, normalized both for gene and genome level for codon and read counts

• ...

• alpha (numeric)dirichlet alpha MOM estimator (imagine mean and variance of probability in 1 value, the lower the value, the higher the variance, mean is decided by the relative value between samples)

• sum_txNorm (integer)total counts per seq normalized per tx

• relative_to_max_score (integer)Percentage use of codon

• type (factor(character))Either "P" or "A"

References

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7196831/

See Also

Other codon: codon_usage(), codon_usage_plot()

Examples

df <- ORFik.template.experiment()[9:10,] # Subset to 2 Ribo-seq libs
## For single library
res <- codon_usage_exp(df, fimport(filepath(df[1,], "pshifted")),
                 min_counts_cds_filter = 10)
# mean_txNorm is adviced scoring column
# codon_usage_plot(res, res$mean_txNorm)
# Default for plot function is the percentage scaled version of mean_txNorm
# codon_usage_plot(res) # This gives check error
## For multiple libs
res2 <- codon_usage_exp(df, outputLibs(df, type = "pshifted", output.mode = "list"),
                 min_counts_cds_filter = 10)
# codon_usage_plot(res2)

Description

Plot codon_usage

Usage

codon_usage_plot(
  res,
  score_column = res$relative_to_max_score,
  ylab = "Ribo-seq library",
  legend.position = "none",
  limit = c(0, max(score_column)),
  midpoint = limit/2,
  monospace_font = TRUE
)

Arguments

Value

a ggplot object

See Also

Other codon: codon_usage(), codon_usage_exp()

Examples

df <- ORFik.template.experiment()[9:10,] # Subset to 2 Ribo-seq libs
## For multiple libs
res2 <- codon_usage_exp(df, outputLibs(df, type = "pshifted", output.mode = "list"),
                 min_counts_cds_filter = 10)
# codon_usage_plot(res2, monospace_font = TRUE) # This gives check error
codon_usage_plot(res2, monospace_font = FALSE) # monospace font looks better

Description

For each unique read in the file, collapse into 1 and state in the fasta header how many reads existed of that type. This is done after trimming usually, works best for reads < 50 read length. Not so effective for 150 bp length mRNA-seq etc.

Usage

collapse.fastq(
  files,
  outdir = file.path(dirname(files[1]), "collapsed"),
  header.out.format = "ribotoolkit",
  compress = FALSE,
  prefix = "collapsed_"
)

Arguments

Value

invisible(NULL), files saved to disc in fasta format.

Examples

fastq.folder <- tempdir() # <- Your fastq files
infiles <- dir(fastq.folder, "*.fastq", full.names = TRUE)
# collapse.fastq(infiles)

Description

For every GRanges, GAlignments read, with the same: seqname, start, (cigar) / width and strand, collapse and give a new meta column called "score", which contains the number of duplicates of that read. If score column already exists, will return input object!

Usage

collapseDuplicatedReads(x, addScoreColumn = TRUE, ...)

Arguments

Value

a GRanges, GAlignments, GAlignmentPairs or data.table object, same as input

Examples

gr <- rep(GRanges("chr1", 1:10,"+"), 2)
collapseDuplicatedReads(gr)

Description

For every GRanges, GAlignments read, with the same: seqname, start, (cigar) / width and strand, collapse and give a new meta column called "score", which contains the number of duplicates of that read. If score column already exists, will return input object!

Usage

## S4 method for signature 'data.table'
collapseDuplicatedReads(
  x,
  addScoreColumn = TRUE,
  addSizeColumn = FALSE,
  reuse.score.column = TRUE,
  keepCigar = FALSE
)

Arguments

Value

a GRanges, GAlignments, GAlignmentPairs or data.table object, same as input

Examples

gr <- rep(GRanges("chr1", 1:10,"+"), 2)
collapseDuplicatedReads(gr)

Description

For every GRanges, GAlignments read, with the same: seqname, start, (cigar) / width and strand, collapse and give a new meta column called "score", which contains the number of duplicates of that read. If score column already exists, will return input object!

Usage

## S4 method for signature 'GAlignmentPairs'
collapseDuplicatedReads(x, addScoreColumn = TRUE)

Arguments

Value

a GRanges, GAlignments, GAlignmentPairs or data.table object, same as input

Examples

gr <- rep(GRanges("chr1", 1:10,"+"), 2)
collapseDuplicatedReads(gr)

Description

For every GRanges, GAlignments read, with the same: seqname, start, (cigar) / width and strand, collapse and give a new meta column called "score", which contains the number of duplicates of that read. If score column already exists, will return input object!

Usage

## S4 method for signature 'GAlignments'
collapseDuplicatedReads(x, addScoreColumn = TRUE, reuse.score.column = TRUE)

Arguments

Value

a GRanges, GAlignments, GAlignmentPairs or data.table object, same as input

Examples

gr <- rep(GRanges("chr1", 1:10,"+"), 2)
collapseDuplicatedReads(gr)

Description

For every GRanges, GAlignments read, with the same: seqname, start, (cigar) / width and strand, collapse and give a new meta column called "score", which contains the number of duplicates of that read. If score column already exists, will return input object!

Usage

## S4 method for signature 'GRanges'
collapseDuplicatedReads(
  x,
  addScoreColumn = TRUE,
  addSizeColumn = FALSE,
  reuse.score.column = TRUE
)

Arguments

Value

a GRanges, GAlignments, GAlignmentPairs or data.table object, same as input

Examples

gr <- rep(GRanges("chr1", 1:10,"+"), 2)
collapseDuplicatedReads(gr)

Description

Given a character vector, get all unique combinations of 2.

Usage

combn.pairs(x)

Arguments

Value

a list of character vector pairs

Examples

df <- ORFik.template.experiment()
ORFik:::combn.pairs(df[, "libtype"])

Description

If you want to get all the NGS and/or sequence features easily, you can use this function. Each feature have a link to an article describing its creation and idea behind it. Look at the functions in the feature family (in the "see also" section below) to see all of them. Example, if you want to know what the "te" column is, check out: ?translationalEff.
A short description of each feature is also shown here:

** NGS features ** If not stated otherwise stated, the feature apply to Ribo-seq.

• countRFP : raw counts of Ribo-seq

• fpkmRFP : FPKM

• fpkmRNA : FPKM of RNA-seq

• te : Translation efficiency Ribo-seq / RNA-seq FPKM

• floss : Fragment length similarity score

• entropyRFP : Positional entropy

• disengagementScores : downstream coverage from ORF

• RRS: Ribosome release score

• RSS: Ribosome staling score

• ORFScores: Periodicity score, does frame 0 have more reads

• ioScore: inside outside score: coverage ORF / coverage rest of transcript

• startCodonCoverage: Coverage over start codon + 2nt before start codon

• startRegionCoverage: Coverage over codon 2 & 3

• startRegionRelative: Peakness of TIS, startCodonCoverage / startRegionCoverage, 0-n

** Sequence features **

• kozak : Similarity to kozak sequence for organism score, 0-1

• gc : GC percentage, 0-1

• StartCodons : Start codon as a string, "ATG"

• StopCodons : stop codon as a string, "TAA"

• fractionLengths : ORF length compared to transcript, 0-1

** uORF features **

• distORFCDS : Distance from ORF stop site to CDS, -n:n

• inFrameCDS : Is ORF in frame with downstream CDS, T/F

• isOverlappingCds : Is ORF overlapping with downstream CDS, T/F

• rankInTx : ORF with most upstream start codon is 1, 1-n

Usage

computeFeatures(
  grl,
  RFP,
  RNA = NULL,
  Gtf,
  faFile = NULL,
  riboStart = 26,
  riboStop = 34,
  sequenceFeatures = TRUE,
  uorfFeatures = TRUE,
  grl.is.sorted = FALSE,
  weight.RFP = 1L,
  weight.RNA = 1L
)

Arguments

Details

If you used CageSeq to reannotate your leaders, your txDB object must contain the reassigned leaders. Use [reassignTxDbByCage()] to get the txdb.
As a note the library is reduced to only reads overlapping 'tx', so the library size in fpkm calculation is done on this subset. This will help remove rRNA and other contaminants.
Also if you have only unique reads with a weight column, explaining the number of duplicated reads, set weights to make calculations correct. See getWeights

Value

a data.table with scores, each column is one score type, name of columns are the names of the scores, i.g [floss()] or [fpkm()]

See Also

Other features: computeFeaturesCage(), countOverlapsW(), disengagementScore(), distToCds(), distToTSS(), entropy(), floss(), fpkm(), fpkm_calc(), fractionLength(), initiationScore(), insideOutsideORF(), isInFrame(), isOverlapping(), kozakSequenceScore(), orfScore(), rankOrder(), ribosomeReleaseScore(), ribosomeStallingScore(), startRegion(), startRegionCoverage(), stopRegion(), subsetCoverage(), translationalEff()

Examples

# Here we make an example from scratch
# Usually the ORFs are found in orfik, which makes names for you etc.
gtf <- system.file("extdata/references/danio_rerio", "annotations.gtf",
 package = "ORFik") ## location of the gtf file

suppressWarnings(txdb <- loadTxdb(gtf))
# use cds' as ORFs for this example
ORFs <- loadRegion(txdb, "cds")
ORFs <- makeORFNames(ORFs) # need ORF names
# make Ribo-seq data,
RFP <- unlistGrl(firstExonPerGroup(ORFs))
computeFeatures(ORFs, RFP, Gtf = txdb)
# For more details see vignettes.

Description

If you have a txdb with correctly reassigned transcripts, use: [computeFeatures()]

Usage

computeFeaturesCage(
  grl,
  RFP,
  RNA = NULL,
  Gtf = NULL,
  tx = NULL,
  fiveUTRs = NULL,
  cds = NULL,
  threeUTRs = NULL,
  faFile = NULL,
  riboStart = 26,
  riboStop = 34,
  sequenceFeatures = TRUE,
  uorfFeatures = TRUE,
  grl.is.sorted = FALSE,
  weight.RFP = 1L,
  weight.RNA = 1L
)

Arguments

Details

A specialized version if you don't have a correct txdb, for example with CAGE reassigned leaders while txdb is not updated. It is 2x faster for tested data. The point of this function is to give you the ability to input transcript etc directly into the function, and not load them from txdb. Each feature have a link to an article describing feature, try ?floss

Value

a data.table with scores, each column is one score type, name of columns are the names of the scores, i.g [floss()] or [fpkm()]

See Also

Other features: computeFeatures(), countOverlapsW(), disengagementScore(), distToCds(), distToTSS(), entropy(), floss(), fpkm(), fpkm_calc(), fractionLength(), initiationScore(), insideOutsideORF(), isInFrame(), isOverlapping(), kozakSequenceScore(), orfScore(), rankOrder(), ribosomeReleaseScore(), ribosomeStallingScore(), startRegion(), startRegionCoverage(), stopRegion(), subsetCoverage(), translationalEff()

Examples

# a small example without cage-seq data:
 # we will find ORFs in the 5' utrs
 # and then calculate features on them
 
 if (requireNamespace("BSgenome.Hsapiens.UCSC.hg19")) {
  library(GenomicFeatures)
  # Get the gtf txdb file
  txdbFile <- system.file("extdata", "hg19_knownGene_sample.sqlite",
  package = "GenomicFeatures")
  txdb <- loadDb(txdbFile)

  # Extract sequences of fiveUTRs.
  fiveUTRs <- fiveUTRsByTranscript(txdb, use.names = TRUE)[1:10]
  faFile <- BSgenome.Hsapiens.UCSC.hg19::Hsapiens
  tx_seqs <- extractTranscriptSeqs(faFile, fiveUTRs)

  # Find all ORFs on those transcripts and get their genomic coordinates
  fiveUTR_ORFs <- findMapORFs(fiveUTRs, tx_seqs)
  unlistedORFs <- unlistGrl(fiveUTR_ORFs)
  # group GRanges by ORFs instead of Transcripts
  fiveUTR_ORFs <- groupGRangesBy(unlistedORFs, unlistedORFs$names)

  # make some toy ribo seq and rna seq data
  starts <- unlistGrl(ORFik:::firstExonPerGroup(fiveUTR_ORFs))
  RFP <- promoters(starts, upstream = 0, downstream = 1)
  score(RFP) <- rep(29, length(RFP)) # the original read widths

  # set RNA seq to duplicate transcripts
  RNA <- unlistGrl(exonsBy(txdb, by = "tx", use.names = TRUE))

  #ORFik:::computeFeaturesCage(grl = fiveUTR_ORFs, RFP = RFP,
  #  RNA = RNA, Gtf = txdb, faFile = faFile)

}
# See vignettes for more examples

Description

Defines a folder for: 1. fastq files (raw data)
2. bam files (processed data)
3. references (organism annotation and STAR index)
4. experiments (Location to store and load all experiment .csv files) Update or use another config using config.save() function.

Usage

config(
  file = config_file(old_config_location = old_config_location),
  old_config_location = "~/Bio_data/ORFik_config.csv"
)

Arguments

Value

a named character vector of length 3

Examples

## Make with default config path
#config()

Description

Get path for ORFik config in cache

Usage

config_file(
  cache = BiocFileCache::getBFCOption("CACHE"),
  query = "ORFik_config",
  ask = interactive(),
  old_config_location = "~/Bio_data/ORFik_config.csv"
)

Arguments

Value

a file path in cache

Examples

config_file()
# Another config path
config_file(query = "ORFik_config_2")

Description

Defines a folder for: 1. fastq files (raw_data)
2. bam files (processed data)
3. references (organism annotation and STAR index)
4. Experiment (name of experiment)

Usage

config.exper(experiment, assembly, type, config = ORFik::config())

Arguments

Value

named character vector of paths for experiment

Examples

# Where should files go in general?
ORFik::config()
# Paths for project: "Alexaki_Human" containing Ribo-seq and RNA-seq:
#config.exper("Alexaki_Human", "Homo_sapiens_GRCh38_101", c("Ribo-seq", "RNA-seq"))

Description

Defines a folder for fastq files (raw_data), bam files (processed data) and references (organism annotation and STAR index)

Usage

config.save(
  file = config_file(),
  fastq.dir = file.path(base.dir, "raw_data"),
  bam.dir = file.path(base.dir, "processed_data"),
  reference.dir = file.path(base.dir, "references"),
  exp.dir = file.path(base.dir, "ORFik_experiments/"),
  base.dir = "~/Bio_data",
  conf = data.frame(type = c("fastq", "bam", "ref", "exp"), directory = c(fastq.dir,
    bam.dir, reference.dir, exp.dir))
)

Arguments

Value

invisible(NULL), file saved to disc

Examples

# Overwrite default config, with new base directory for files
#config.save(base.dir = "/media/Bio_data/") # Output files go here instead
# of ~/Bio_data
## Dont do this, but for understanding here is how to make a second config
#new_config_path <- config_file(query = "ORFik_config_2")
#config.save(new_config_path, "/media/Bio_data/raw_data/",
# "/media/Bio_data/processed_data", /media/Bio_data/references/)

Description

Saved by default in folder "ofst" relative to default libraries of experiment. Speeds up loading of full files compared to bam by large margins.

Usage

convert_bam_to_ofst(
  df,
  in_files = filepath(df, "default"),
  out_dir = file.path(libFolder(df), "ofst"),
  verbose = TRUE,
  strandMode = rep(0, length(in_files))
)

Arguments

Details

If you want to keep bam files loaded or faster conversion if you already have them loaded, use ORFik::convertLibs instead

Value

invisible(NULL), files saved to disc

See Also

Other lib_converters: convertLibs(), convert_to_bigWig(), convert_to_covRle(), convert_to_covRleList()

Examples

df <- ORFik.template.experiment.zf()
## Usually do default folder, here we use tmpdir
folder_to_save <- file.path(tempdir(), "ofst")
convert_bam_to_ofst(df, out_dir = folder_to_save)
fimport(file.path(folder_to_save, "ribo-seq.ofst"))

Description

Convert to BigWig

Usage

convert_to_bigWig(
  df,
  in_files = filepath(df, "pshifted"),
  out_dir = file.path(libFolder(df), "bigwig"),
  split.by.strand = TRUE,
  split.by.readlength = FALSE,
  seq_info = seqinfo(df),
  weight = "score",
  is_pre_collapsed = FALSE,
  verbose = TRUE
)

Arguments

Value

invisible(NULL), files saved to disc

See Also

Other lib_converters: convertLibs(), convert_bam_to_ofst(), convert_to_covRle(), convert_to_covRleList()

Examples

df <- ORFik.template.experiment()[10,]
## Usually do default folder, here we use tmpdir
folder_to_save <- file.path(tempdir(), "bigwig")
convert_to_bigWig(df, out_dir = folder_to_save)
fimport(file.path(folder_to_save, c("RFP_Mutant_rep2_forward.bigWig",
 "RFP_Mutant_rep2_reverse.bigWig")))

Description

Saved by default in folder "cov_RLE" relative to default libraries of experiment

Usage

convert_to_covRle(
  df,
  in_files = filepath(df, "pshifted"),
  out_dir = file.path(libFolder(df), "cov_RLE"),
  split.by.strand = TRUE,
  split.by.readlength = FALSE,
  seq_info = seqinfo(df),
  weight = "score",
  verbose = TRUE
)

Arguments

Value

invisible(NULL), files saved to disc

See Also

Other lib_converters: convertLibs(), convert_bam_to_ofst(), convert_to_bigWig(), convert_to_covRleList()

Examples

df <- ORFik.template.experiment()[10,]
## Usually do default folder, here we use tmpdir
folder_to_save <- file.path(tempdir(), "cov_RLE")
convert_to_covRle(df, out_dir = folder_to_save)
fimport(file.path(folder_to_save, "RFP_Mutant_rep2.covrds"))

Description

Useful to store reads separated by readlength, for much faster coverage calculation. Saved by default in folder "cov_RLE_List" relative to default libraries of experiment

Usage

convert_to_covRleList(
  df,
  in_files = filepath(df, "pshifted"),
  out_dir = file.path(libFolder(df), "cov_RLE_List"),
  out_dir_merged = file.path(libFolder(df), "cov_RLE"),
  split.by.strand = TRUE,
  seq_info = seqinfo(df),
  weight = "score",
  verbose = TRUE
)

Arguments

Value

invisible(NULL), files saved to disc

See Also

Other lib_converters: convertLibs(), convert_bam_to_ofst(), convert_to_bigWig(), convert_to_covRle()

Examples

df <- ORFik.template.experiment()[10,]
## Usually do default folder, here we use tmpdir
folder_to_save <- file.path(tempdir(), "cov_RLE_List")
folder_to_save_merged <- file.path(tempdir(), "cov_RLE")
ORFik:::convert_to_covRleList(df, out_dir = folder_to_save,
out_dir_merged = folder_to_save_merged)
fimport(file.path(folder_to_save, "RFP_Mutant_rep2.covrds"))

Description

Will split files by chromosome for faster loading for now. This feature might change in the future!

Usage

convert_to_fstWig(
  df,
  in_files = filepath(df, "pshifted"),
  out_dir = file.path(libFolder(df), "fstwig"),
  split.by.strand = TRUE,
  split.by.readlength = FALSE,
  seq_info = seqinfo(df),
  weight = "score",
  is_pre_collapsed = FALSE,
  verbose = TRUE
)

Arguments

Value

invisible(NULL), files saved to disc

Description

Export as either .ofst, .wig, .bigWig,.bedo (legacy format) or .bedoc (legacy format) files:
Export files as .ofst for fastest load speed into R.
Export files as .wig / bigWig for use in IGV or other genome browsers.
The input files are checked if they exist from: envExp(df).

Usage

convertLibs(
  df,
  out.dir = libFolder(df),
  addScoreColumn = TRUE,
  addSizeColumn = TRUE,
  must.overlap = NULL,
  method = "None",
  type = "ofst",
  input.type = "ofst",
  reassign.when.saving = FALSE,
  envir = envExp(df),
  force = TRUE,
  library.names = bamVarName(df),
  libs = outputLibs(df, type = input.type, chrStyle = must.overlap, library.names =
    library.names, output.mode = "list", force = force, BPPARAM = BPPARAM),
  BPPARAM = bpparam()
)

Arguments

Details

We advice you to not use this directly, as other function are more safe for library type conversions. See family description below. This is mostly used internally in ORFik. It is only adviced to use if large bam files are already loaded in R and conversions are wanted from those.

See export.ofst, export.wiggle, export.bedo and export.bedoc for information on file formats.
If libraries of the experiment are already loaded into environment (default: .globalEnv) is will export using those files as templates. If they are not in environment the .ofst files from the bam files are loaded (unless you are converting to .ofst then the .bam files are loaded).

Value

invisible NULL (saves files to disc or R .GlobalEnv)

See Also

Other lib_converters: convert_bam_to_ofst(), convert_to_bigWig(), convert_to_covRle(), convert_to_covRleList()

Examples

df <- ORFik.template.experiment()
#convertLibs(df, out.dir = NULL)
# Keep only 5' ends of reads
#convertLibs(df, out.dir = NULL, method = "5prime")

Description

There are 5 ways of doing this
1. Take 5' ends, reduce away rest (5prime)
2. Take 3' ends, reduce away rest (3prime)
3. Tile to 1-mers and include all (tileAll)
4. Take middle point per GRanges (middle)
5. Get original with metacolumns (None)
You can also do multiple at a time, then output is GRangesList, where each list group is the operation (5prime is [1], 3prime is [2] etc)
Many other ways to do this have their own functions, like startSites and stopSites etc. To retain information on original width, set addSizeColumn to TRUE. To compress data, 1 GRanges object per unique read, set addScoreColumn to TRUE. This will give you a score column with how many duplicated reads there were in the specified region.

Usage

convertToOneBasedRanges(
  gr,
  method = "5prime",
  addScoreColumn = FALSE,
  addSizeColumn = FALSE,
  after.softclips = TRUE,
  along.reference = FALSE,
  reuse.score.column = TRUE
)

Arguments

Details

NOTE: Note: For cigar based ranges (GAlignments), the 5' end is the first non clipped base (neither soft clipped or hard clipped from 5'). This is following the default of bioconductor. For special case of GAlignmentPairs, 5prime will only use left (first) 5' end and read and 3prime will use only right (last) 3' end of read in pair. tileAll and middle can possibly find poinst that are not in the reads since: lets say pair is 1-5 and 10-15, middle is 7, which is not in the read.

Value

Converted GRanges object

See Also

Other utils: bedToGR(), export.bed12(), export.bigWig(), export.fstwig(), export.wiggle(), fimport(), findFa(), fread.bed(), optimizeReads(), readBam(), readBigWig(), readWig()

Examples

gr <- GRanges("chr1", 1:10,"+")
# 5 prime ends
convertToOneBasedRanges(gr)
# is equal to convertToOneBasedRanges(gr, method = "5prime")
# 3 prime ends
convertToOneBasedRanges(gr, method = "3prime")
# With lengths
convertToOneBasedRanges(gr, addSizeColumn = TRUE)
# With score (# of replicates)
gr <- rep(gr, 2)
convertToOneBasedRanges(gr, addSizeColumn = TRUE, addScoreColumn = TRUE)

Description

Get correlation between columns

Usage

cor_plot(
  dt_cor,
  col = c(low = "blue", high = "red", mid = "white", na.value = "white"),
  limit = c(ifelse(min(dt_cor$Cor, na.rm = TRUE) < 0, -1, 0), 1),
  midpoint = mean(limit),
  label_name = "Pearson\nCorrelation",
  text_size = 4,
  legend.position = c(0.4, 0.7),
  legend.direction = "horizontal"
)

Arguments

Value

a ggplot (heatmap)

Description

Get correlation between columns

Usage

cor_table(
  dt,
  method = c("pearson", "spearman")[1],
  upper_triangle = TRUE,
  decimals = 2,
  melt = TRUE,
  na.rm.melt = TRUE
)

Arguments

Value

a data.table with 3 columns, Var1, Var2 and Cor

Description

Get correlation plot of raw counts and/or log2(count + 1) over selected region in: c("mrna", "leaders", "cds", "trailers")

Note on correlation: Pearson correlation, using pairwise observations to fill in NA values for the covariance matrix.

Usage

correlation.plots(
  df,
  output.dir,
  region = "mrna",
  type = "fpkm",
  height = 400,
  width = 400,
  size = 0.15,
  plot.ext = ".pdf",
  complex.correlation.plots = TRUE,
  data_for_pairs = countTable(df, region, type = type),
  as_gg_list = FALSE,
  text_size = 4,
  method = c("pearson", "spearman")[1]
)

Arguments

Value

invisible(NULL) / if as_gg_list is TRUE, return a list of raw plots.

Description

Similar to countOverlaps, but takes an optional weight column. This is usually the score column

Usage

countOverlapsW(query, subject, weight = NULL, ...)

Arguments

Value

a named vector of number of overlaps to subject weigthed by 'weight' column.

See Also

Other features: computeFeatures(), computeFeaturesCage(), disengagementScore(), distToCds(), distToTSS(), entropy(), floss(), fpkm(), fpkm_calc(), fractionLength(), initiationScore(), insideOutsideORF(), isInFrame(), isOverlapping(), kozakSequenceScore(), orfScore(), rankOrder(), ribosomeReleaseScore(), ribosomeStallingScore(), startRegion(), startRegionCoverage(), stopRegion(), subsetCoverage(), translationalEff()

Examples

gr1 <- GRanges(seqnames="chr1",
               ranges=IRanges(start = c(4, 9, 10, 30),
                              end = c(4, 15, 20, 31)),
               strand="+")
gr2 <- GRanges(seqnames="chr1",
               ranges=IRanges(start = c(1, 4, 15, 25),
                              end = c(2, 4, 20, 26)),
               strand=c("+"),
               score=c(10, 20, 15, 5))
countOverlaps(gr1, gr2)
countOverlapsW(gr1, gr2, weight = "score")

Description

Used to quickly load pre-created read count tables to R.
If df is experiment: Extracts by getting /QC_STATS directory, and searching for region Requires ORFikQC to have been run on experiment, to get default count tables!

Usage

countTable(
  df,
  region = "mrna",
  type = "count",
  collapse = FALSE,
  count.folder = "default"
)

Arguments

Details

If df is path to folder: Loads the the file in that directory with the regex region.rds, where region is what is defined by argument, if multiple exist, see if any start with "countTable_", if so, subset. If loaded as SummarizedExperiment or deseq, the colData will be made from ORFik.experiment information.

Value

a data.table/SummarizedExperiment/DESeq object of columns as counts / normalized counts per library, column name is name of library. Rownames must be unique for now. Might change.

See Also

Other countTable: countTable_regions()

Examples

# Make experiment
df <- ORFik.template.experiment()
# Make QC report to get counts ++ (not needed for this template)
# ORFikQC(df)

# Get count Table of mrnas
# countTable(df, "mrna")
# Get count Table of cds
# countTable(df, "cds")
# Get count Table of mrnas as fpkm values
# countTable(df, "mrna", type = "count")
# Get count Table of mrnas with collapsed replicates
# countTable(df, "mrna", collapse = TRUE)
# Get count Table of mrnas as summarizedExperiment
# countTable(df, "mrna", type = "summarized")
# Get count Table of mrnas as DESeq2 object,
# for differential expression analysis
# countTable(df, "mrna", type = "deseq")

Description

By default will make count tables over mRNA, leaders, cds and trailers for all libraries in experiment. region

Usage

countTable_regions(
  df,
  out.dir = libFolder(df),
  longestPerGene = FALSE,
  geneOrTxNames = "tx",
  regions = c("mrna", "leaders", "cds", "trailers"),
  type = "count",
  lib.type = "ofst",
  weight = "score",
  rel.dir = "QC_STATS",
  forceRemake = FALSE,
  library.names = bamVarName(df),
  BPPARAM = bpparam()
)

Arguments

Value

a list of data.table, 1 data.table per region. The regions will be the names the list elements.

See Also

Other countTable: countTable()

Examples

##Make experiment
df <- ORFik.template.experiment()
## Create count tables for all default regions
# countTable_regions(df)
## Pshifted reads (first create pshiftead libs)
# countTable_regions(df, lib.type = "pshifted", rel.dir = "pshifted")

Description

Convert coverage RleList to data.table

Usage

coverage_to_dt(
  coverage,
  keep.names = TRUE,
  withFrames = FALSE,
  weight = "score",
  drop.zero.dt = FALSE,
  fraction = NULL
)

Arguments

Value

a data.table with column names c("count" [numeric or integer], "genes" [integer], "position" [integer])

Description

Extends the function with direct genome coverage input, see coverageByTranscript for original function.

Usage

coverageByTranscriptC(x, transcripts, ignore.strand = !strandMode(x))

Arguments

Value

Integer Rle of coverage, 1 per transcript

Description

Extends the function with weights, see coverageByTranscript for original function.

Usage

coverageByTranscriptW(
  x,
  transcripts,
  ignore.strand = FALSE,
  weight = 1L,
  seqinfo.x.is.correct = FALSE
)

Arguments

Value

Integer Rle of coverage, 1 per transcript

Description

Creates a ggplot representing a heatmap of coverage:

• Rows : Position in region

• Columns : Read length

• Index intensity : (color) coverage scoring per index.

Coverage rows in heat map is fraction, usually fractions is divided into unique read lengths (standard Illumina is 76 unique widths, with some minimum cutoff like 15.) Coverage column in heat map is score, default zscore of counts. These are the relative positions you are plotting to. Like +/- relative to TIS or TSS.

Usage

coverageHeatMap(
  coverage,
  output = NULL,
  scoring = "zscore",
  legendPos = "right",
  addFracPlot = FALSE,
  xlab = "Position relative to start site",
  ylab = "Protected fragment length",
  colors = "default",
  title = NULL,
  increments.y = "auto",
  gradient.max = max(coverage$score)
)

Arguments

Details

Colors: Remember if you want to change anything like colors, just return the ggplot object, and reassign like: obj + scale_color_brewer() etc. Standard colors are:

• 0 reads in whole readlength :gray

• few reads in position :white

• medium reads in position :yellow

• many reads in position :dark blue

Value

a ggplot object of the coverage plot, NULL if output is set, then the plot will only be saved to location.

See Also

Other heatmaps: heatMapL(), heatMapRegion(), heatMap_single()

Other coveragePlot: pSitePlot(), savePlot(), windowCoveragePlot()

Examples

# An ORF
grl <- GRangesList(tx1 = GRanges("1", IRanges(1, 6), "+"))
# Ribo-seq reads
range <- IRanges(c(rep(1, 3), 2, 3, rep(4, 2), 5, 6), width = 1 )
reads <- GRanges("1", range, "+")
reads$size <- c(rep(28, 5), rep(29, 4)) # read size
coverage <- windowPerReadLength(grl, reads = reads, upstream = 0,
                                downstream = 5)

coverageHeatMap(coverage)

# With top sum bar
coverageHeatMap(coverage, addFracPlot = TRUE)
# See vignette for more examples

Description

It tiles each GRangesList group to width 1, and finds hits per position.
A range from 1:5 will split into c(1,2,3,4,5) and count hits on each. This is a safer speedup of coverageByTranscript from GenomicFeatures. It also gives the possibility to return as data.table, for faster computations.

Usage

coveragePerTiling(
  grl,
  reads,
  is.sorted = FALSE,
  keep.names = TRUE,
  as.data.table = FALSE,
  withFrames = FALSE,
  weight = "score",
  drop.zero.dt = FALSE,
  fraction = NULL
)

Arguments

Details

NOTE: If reads contains a $score column, it will presume that this is the number of replicates per reads, weights for the coverage() function. So delete the score column or set weight to something else if this is not wanted.

Value

a numeric RleList, one numeric-Rle per group with # of hits per position. Or data.table if as.data.table is TRUE, with column names c("count" [numeric or integer], "genes" [integer], "position" [integer])

See Also

Other ExtendGenomicRanges: asTX(), extendLeaders(), extendTrailers(), reduceKeepAttr(), tile1(), txSeqsFromFa(), windowPerGroup()

Examples

ORF <- GRanges(seqnames = "1",
               ranges = IRanges(start = c(1, 10, 20),
                                end = c(5, 15, 25)),
               strand = "+")
grl <- GRangesList(tx1_1 = ORF)
RFP <- GRanges("1", IRanges(25, 25), "+")
coveragePerTiling(grl, RFP, is.sorted = TRUE)
# now as data.table with frames
coveragePerTiling(grl, RFP, is.sorted = TRUE, as.data.table = TRUE,
                  withFrames = TRUE)
# With score column (usually replicated reads on that position)
RFP <- GRanges("1", IRanges(25, 25), "+", score = 5)
dt <- coveragePerTiling(grl, RFP, is.sorted = TRUE,
                        as.data.table = TRUE, withFrames = TRUE)
class(dt$count) # numeric
# With integer score column (faster and less space usage)
RFP <- GRanges("1", IRanges(25, 25), "+", score = 5L)
dt <- coveragePerTiling(grl, RFP, is.sorted = TRUE,
                        as.data.table = TRUE, withFrames = TRUE)
class(dt$count) # integer

Description

Different scorings and groupings of a coverage representation.

Usage

coverageScorings(coverage, scoring = "zscore", copy.dt = TRUE)

Arguments

Details

Usually output of metaWindow or scaledWindowPositions is input in this function.

Content of coverage data.table: It must contain the count and position columns.

genes column: If you have multiple windows, the genes column must define which gene/transcript grouping the different counts belong to. If there is only a meta window or only 1 gene/transcript, then this column is not needed.

fraction column: If you have coverage of i.e RNA-seq and Ribo-seq, or TCP -seq of large and small subunite, divide into fractions. Like factor(RNA, RFP)

feature column: If gene group is subdivided into parts, like gene is transcripts, and feature column can be c(leader, cds, trailer) etc.

Given a data.table coverage of counts, add a scoring scheme. per: the grouping given, if genes is defined, group by per gene in default scoring.
Scorings:

• zscore (count-windowMean)/windowSD per)

• transcriptNormalized (sum(count / sum of counts per))

• mean (mean(count per))

• median (median(count per))

• sum (count per)

• log2sum (count per)

• log10sum (count per)

• sumLength (count per) / number of windows

• meanPos (mean per position per gene) used in scaledWindowPositions

• sumPos (sum per position per gene) used in scaledWindowPositions

• frameSum (sum per frame per gene) used in ORFScore

• frameSumPerL (sum per frame per read length)

• frameSumPerLG (sum per frame per read length per gene)

• fracPos (fraction of counts per position per gene)

• periodic (Fourier transform periodicity of meta coverage per fraction)

• NULL (no grouping, return input directly)

Value

a data.table with new scores (size dependent on score used)

See Also

Other coverage: metaWindow(), regionPerReadLength(), scaledWindowPositions(), windowPerReadLength()

Examples

dt <- data.table::data.table(count = c(4, 1, 1, 4, 2, 3),
                             position = c(1, 2, 3, 4, 5, 6))
coverageScorings(dt, scoring = "zscore")

# with grouping gene
dt$genes <- c(rep("tx1", 3), rep("tx2", 3))
coverageScorings(dt, scoring = "zscore")

Description

Coverage Rlelist for both strands

Usage

covRle(forward = RleList(), reverse = RleList())

Arguments

Value

a covRle object

See Also

Other covRLE: covRle-class, covRleFromGR(), covRleList, covRleList-class

Examples

covRle()
covRle(RleList(), RleList())
chr_rle <- RleList(chr1 = Rle(c(1,2,3), c(1,2,3)))
covRle(chr_rle, chr_rle)

Description

Given a run of coverage(x) where x are reads, this class combines the 2 strands into 1 object

Value

a covRLE object

See Also

Other covRLE: covRle, covRleFromGR(), covRleList, covRleList-class

Description

Convert GRanges to covRle

Usage

covRleFromGR(x, weight = "AUTO", ignore.strand = FALSE)

Arguments

Value

covRle object

See Also

Other covRLE: covRle, covRle-class, covRleList, covRleList-class

Examples

seqlengths <- as.integer(c(200, 300))
names(seqlengths) <- c("chr1", "chr2")
gr <- GRanges(seqnames = c("chr1", "chr1", "chr2", "chr2"),
               ranges = IRanges(start = c(10, 50, 100, 150), end = c(40, 80, 129, 179)),
               strand = c("+", "+", "-", "-"), seqlengths = seqlengths)
cov_both_strands <- covRleFromGR(gr)
cov_both_strands
cov_ignore_strand <- covRleFromGR(gr, ignore.strand = TRUE)
cov_ignore_strand
strandMode(cov_both_strands)
strandMode(cov_ignore_strand)

Description

Coverage Rlelist for both strands

Usage

covRleList(list, fraction = names(list))

Arguments

Value

a covRleList object

See Also

Other covRLE: covRle, covRle-class, covRleFromGR(), covRleList-class

Examples

covRleList(List(covRle()))

Description

Given a run of coverage(x) where x are reads, this covRle combines the 2 strands into 1 object This list can again combine these into 1 object, with accession functions and generalizations.

Value

a covRleList object

See Also

Other covRLE: covRle, covRle-class, covRleFromGR(), covRleList

Create an ORFik experiment

Description

Create a single R object that stores and controls all results relevant to a specific Next generation sequencing experiment. Click the experiment link above in the title if you are not sure what an ORFik experiment is.

By using files in a folder / folders. It will make an experiment table with information per sample, this object allows you to use the extensive API in ORFik that works on experiments.

Information Auto-detection:
There will be several columns you can fill in, when creating the object, if the files have logical names like (RNA-seq_WT_rep1.bam) it will try to auto-detect the most likely values for the columns. Like if it is RNA-seq or Ribo-seq, Wild type or mutant, is this replicate 1 or 2 etc.
You will have to fill in the details that were not auto detected. Easiest way to fill in the blanks are in a csv editor like libre Office or excel. You can also remake the experiment and specify the specific column manually. Remember that each row (sample) must have a unique combination of values. An extra column called "reverse" is made if there are paired data, like +/- strand wig files.

Usage

create.experiment(
  dir,
  exper,
  saveDir = ORFik::config()["exp"],
  txdb = "",
  fa = "",
  organism = "",
  assembly = "",
  pairedEndBam = FALSE,
  viewTemplate = FALSE,
  types = c("bam", "bed", "wig", "ofst"),
  libtype = "auto",
  stage = "auto",
  rep = "auto",
  condition = "auto",
  fraction = "auto",
  author = "",
  files = findLibrariesInFolder(dir, types, pairedEndBam),
  result_folder = NULL,
  runIDs = extract_run_id(files)
)

Arguments

Value

a data.frame, NOTE: this is not a ORFik experiment, only a template for it!

See Also

Other ORFik_experiment: ORFik.template.experiment(), ORFik.template.experiment.zf(), bamVarName(), experiment-class, filepath(), libraryTypes(), organism,experiment-method, outputLibs(), read.experiment(), save.experiment(), validateExperiments()

Examples

# 1. Pick directory
dir <- system.file("extdata/Homo_sapiens_sample", "", package = "ORFik")
# 2. Pick an experiment name
exper <- "ORFik"
# 3. Pick .gff/.gtf location
txdb <- system.file("extdata/references/homo_sapiens",
                    "Homo_sapiens_dummy.gtf.db", package = "ORFik")
# 4. Pick fasta genome of organism
fa <- system.file("extdata/references/homo_sapiens",
                  "Homo_sapiens_dummy.fasta", package = "ORFik")
# 5. Set organism (optional)
org <- "Homo sapiens"

# Create temple not saved on disc yet:
template <- create.experiment(dir = dir, exper, txdb = txdb,
                              saveDir = NULL,
                              fa = fa, organism = org,
                              viewTemplate = FALSE)
## Now fix non-unique rows: either is libre office, microsoft excel, or in R
template$X5[6] <- "heart"
# read experiment (if you set correctly)
df <- read.experiment(template)
# Save with: save.experiment(df, file = "path/to/save/experiment.csv")

## Create and save experiment directly:
## Default location of experiments is ORFik::config()["exp"]
#template <- create.experiment(dir = dir, exper, txdb = txdb,
#                               fa = fa, organism = org,
#                               viewTemplate = FALSE)
## Custom location (If you work in a team, use a shared folder)
#template <- create.experiment(dir = dir, exper, txdb = txdb,
#                               saveDir = "~/MY/CUSTOME/LOCATION",
#                               fa = fa, organism = org,
#                               viewTemplate = FALSE)

Description

Creates GRanges object as a trailer for ORFranges representing ORF, maintaining restrictions of transcriptRanges. Assumes that ORFranges is on the transcriptRanges, strands and seqlevels are in agreement. When lengthOFtrailer is smaller than space left on the transcript than all available space is returned as trailer.

Usage

defineTrailer(ORFranges, transcriptRanges, lengthOftrailer = 200)

Arguments

Details

It assumes that ORFranges and transcriptRanges are not sorted when on minus strand. Should be like: (200, 600) (50, 100)

Value

A GRanges object of trailer.

See Also

Other ORFHelpers: longestORFs(), mapToGRanges(), orfID(), startCodons(), startSites(), stopCodons(), stopSites(), txNames(), uniqueGroups(), uniqueOrder()

Examples

ORFranges <- GRanges(seqnames = Rle(rep("1", 3)),
                     ranges = IRanges(start = c(1, 10, 20),
                                      end = c(5, 15, 25)),
                     strand = "+")
transcriptRanges <- GRanges(seqnames = Rle(rep("1", 5)),
                            ranges = IRanges(start = c(1, 10, 20, 30, 40),
                                             end = c(5, 15, 25, 35, 45)),
                     strand = "+")
defineTrailer(ORFranges, transcriptRanges)

Description

This is the preparation step of DESeq2 analysis using ORFik::DEG.analysis. It is exported so that you can do this step in standalone, usually you want to use DEG.analysis directly.

Usage

DEG_model(
  df,
  target.contrast = design[1],
  design = ORFik::design(df),
  p.value = 0.05,
  counts = countTable(df, "mrna", type = "summarized"),
  batch.effect = TRUE
)

Arguments

Value

a DESeqDataSet object with results stored as metadata columns.

See Also

Other DifferentialExpression: DEG.plot.static(), DTEG.analysis(), DTEG.plot(), te.table(), te_rna.plot()

Examples

## Simple example (use ORFik template, then use only RNA-seq)
df <- ORFik.template.experiment()
df.rna <- df[df$libtype == "RNA",]
design(df.rna) # The full experimental design
target.contrast <- design(df.rna)[1] # Default target contrast
#ddsMat_rna <- DEG_model(df.rna, target.contrast)

Description

Get DESeq2 model results from DESeqDataSet

Usage

DEG_model_results(ddsMat_rna, target.contrast, pairs, p.value = 0.05)

Arguments

Value

a data.table

Examples

## Simple example (use ORFik template, then use only RNA-seq)
df <- ORFik.template.experiment()
df.rna <- df[df$libtype == "RNA",]
design(df.rna) # The full experimental design
target.contrast <- design(df.rna)[1] # Default target contrast
#ddsMat_rna <- DEG_model(df.rna, target.contrast)
#pairs <- combn.pairs(unlist(df[, target.contrast]))
#dt <- DEG_model_results(ddsMat_rna, target.contrast, pairs)

Description

If you do not have a valid DESEQ2 experimental setup (contrast), you can use this simplified test

Usage

DEG_model_simple(
  df,
  target.contrast = design[1],
  design = ORFik::design(df),
  p.value = 0.05,
  counts = countTable(df, "mrna", type = "summarized"),
  batch.effect = FALSE
)

Arguments

Value

a data.table of fpkm ratios

Examples

## Simple example (use ORFik template, then use only RNA-seq)
df <- ORFik.template.experiment()
df <- df[df$libtype == "RNA",]
#dt <- DEG_model_simple(df)

Description

Expression analysis of 1 dimension, usually between conditions of RNA-seq.
Using the standardized DESeq2 pipeline flow.
Creates a DESeq model (given x is the target.contrast argument) (usually 'condition' column)
1. RNA-seq model: design = ~ x (differences between the x groups in RNA-seq)

Usage

DEG.analysis(
  df,
  target.contrast = design[1],
  design = ORFik::design(df),
  p.value = 0.05,
  counts = countTable(df, "mrna", type = "summarized"),
  batch.effect = TRUE,
  pairs = combn.pairs(unlist(df[, target.contrast]))
)

Arguments

Details

#' Analysis is done between each possible combination of levels in the target contrast If target contrast is the condition column, with factor levels: WT, mut1 and mut2 with 3 replicates each. You get comparison of WT vs mut1, WT vs mut2 and mut1 vs mut2.
The respective result categories are defined as: (given a user defined p value, shown here as 0.05):
Significant - p-value adjusted < 0.05 (p-value cutoff decided by 'p.value argument)

The LFC values are shrunken by lfcShrink(type = "normal").

Remember that DESeq by default can not do global change analysis, it can only find subsets with changes in LFC!

Value

a data.table with columns: (contrast variable, gene id, regulation status, log fold changes, p.adjust values, mean counts)

References

doi: 10.1002/cpmb.108

See Also

Other DifferentialExpression: DEG.plot.static(), DEG_model(), DTEG.plot(), te.table(), te_rna.plot()

Examples

## Simple example (use ORFik template, then use only RNA-seq)
df <- ORFik.template.experiment()
df.rna <- df[df$libtype == "RNA",]
design(df.rna) # The full experimental design
design(df.rna)[1] # Default target contrast
#dt <- DEG.analysis(df.rna)

Description

Plot setup:
X-axis: mean counts Y-axis: Log2 fold changes For explanation of plot, see DEG.analysis

Usage

DEG.plot.static(
  dt,
  output.dir = NULL,
  p.value.label = 0.05,
  plot.title = "",
  plot.ext = ".pdf",
  width = 6,
  height = 6,
  dot.size = 0.4,
  xlim = "auto",
  ylim = "bidir.max",
  relative.name = paste0("DEG_plot", plot.ext)
)

Arguments

Value

a ggplot object

See Also

Other DifferentialExpression: DEG_model(), DTEG.analysis(), DTEG.plot(), te.table(), te_rna.plot()

Examples

df <- ORFik.template.experiment()
df.rna <- df[df$libtype == "RNA",]
#dt <- DEG.analysis(df.rna)
#Default scaling
#DEG.plot.static(dt)
#Manual scaling
#DEG.plot.static(dt, xlim = c(-2, 2), ylim = c(-2, 2))

Description

Get experimental design Find the column/columns that create a separation between samples, by default skips replicate and choose first that is from either: libtype, condition, stage and fraction.

Usage

## S4 method for signature 'experiment'
design(
  object,
  batch.correction.design = FALSE,
  as.formula = FALSE,
  multi.factor = TRUE
)

Arguments

Value

a character (name of column) or a formula

Examples

df <- ORFik.template.experiment()
design(df) # The 2 columns that decides the design here
# If we subset it changes
design(df[df$libtype == "RFP",])
# Only single factor design, it picks first
design(df, multi.factor = FALSE)

Description

Finding all ORFs: 1. Find all ORFs in mRNA using ORFik findORFs, with defined parameters.
To create the candidate ORFs (all ORFs returned):
Steps (candidate set):
Define a candidate search set by these 3 rules:
1.a Allowed ORF type: uORF, NTE, etc (only keep these in candidate list)
1.b Must have at least x reads over whole orf (default 10 reads)
1.c Must have at least x reads over start site (default 3 reads)
The total list is defined by these names, and saved according to allowed ORF type/types.
To create the prediction status (TRUE/FALSE) per candidate
Steps (prediction status)
(UP_NT is a 20nt window upstream of ORF, that stops 2NT before ORF starts) :
1. ORF mean reads per NT > (UP_NT mean reads per NT * 1.3)
2. ORFScore > 2.5
3. TIS total reads + 3 > ORF median reads per NT
4. Given expression above, a TRUE prediction is defined with the AND operatior: 1. & 2. & 3.

In code that is:
predicted <- (orfs_cov_stats$mean > upstream_cov_stats$mean*1.3) & orfs_cov_stats$ORFScores > 2.5 & ((reads_start[candidates] + 3) > orfs_cov_stats$median)

Usage

detect_ribo_orfs(
  df,
  out_folder,
  ORF_categories_to_keep,
  prefix_result = paste(c(ORF_categories_to_keep, gsub(" ", "_", organism(df))), collapse
    = "_"),
  mrna = loadRegion(df, "mrna"),
  cds = loadRegion(df, "cds"),
  libraries = outputLibs(df, type = "pshifted", output = "envirlist"),
  orf_candidate_ranges = findORFs(seqs = txSeqsFromFa(mrna, df, TRUE), longestORF =
    longestORF, startCodon = startCodon, stopCodon = stopCodon, minimumLength =
    minimumLength),
  export_metrics_table = TRUE,
  longestORF = FALSE,
  startCodon = startDefinition(1),
  stopCodon = stopDefinition(1),
  minimumLength = 0,
  minimum_reads_ORF = 10,
  minimum_reads_start = 3
)

Arguments

Value

invisible(NULL), all ORF results saved to disc

Examples

# Pre requisites
# 1. Create ORFik experiment
#  ORFik::create.experiment(...)
# 2. Create ORFik optimized annotation:
# makeTxdbFromGenome(gtf = ORFik:::getGtfPathFromTxdb(df), genome = df@fafile,
#                      organism = organism(df), optimize = TRUE)
# 3. There must exist pshifted reads, either as default files, or in a relative folder called
# "./pshifted/". See ?shiftFootprintsByExperiment
# EXAMPLE:
df <- ORFik.template.experiment()
df <- df[df$libtype == "RFP",][c(1,2),]
result_folder <- riboORFsFolder(df, tempdir())
results <- detect_ribo_orfs(df, result_folder, c("uORF", "uoORF", "annotated", "NTE"))

# Load results of annotated ORFs
table <- riboORFs(df[1,], type = "table", result_folder)
table # See all statistics
sum(table$predicted) # How many were predicted as Ribo-seq ORFs
# Load 2 results
table <- riboORFs(df[1:2,], type = "table", result_folder)
table # See all statistics
sum(table$predicted) # How many were predicted as Ribo-seq ORFs

# Load GRangesList
candidates_gr <- riboORFs(df[1,], type = "ranges_candidates", result_folder)
prediction <- riboORFs(df[1,], type = "predictions", result_folder)

predicted_gr <- riboORFs(df[1:2,], type = "ranges_predictions", result_folder)
identical(predicted_gr[[1]], candidates_gr[[1]][prediction[[1]]])
## Inspect predictions in RiboCrypt
# library(RiboCrypt)
# Inspect Predicted
view <- predicted_gr[[1]][1]
#multiOmicsPlot_ORFikExp(view, df, view, leader_extension = 100, trailer_extension = 100)
# Inspect not predicted
view <- candidates_gr[[1]][!prediction[[1]]][1]
#multiOmicsPlot_ORFikExp(view, df, view, leader_extension = 100, trailer_extension = 100)

Description

Utilizes periodicity measurement (Fourier transform), and change point analysis to detect ribosomal footprint shifts for each of the ribosomal read lengths. Returns subset of read lengths and their shifts for which top covered transcripts follow periodicity measure. Each shift value assumes 5' anchoring of the reads, so that output offsets values will shift 5' anchored footprints to be on the p-site of the ribosome. The E-site will be shift + 3 and A site will be shift - 3. So update to these, if you rather want those.

Usage

detectRibosomeShifts(
  footprints,
  txdb,
  start = TRUE,
  stop = FALSE,
  top_tx = 10L,
  minFiveUTR = 30L,
  minCDS = 150L,
  minThreeUTR = if (stop) {
     30
 } else NULL,
  txNames = filterTranscripts(txdb, minFiveUTR, minCDS, minThreeUTR),
  firstN = 150L,
  tx = NULL,
  min_reads = 1000,
  min_reads_TIS = 50,
  accepted.lengths = 26:34,
  heatmap = FALSE,
  must.be.periodic = TRUE,
  strict.fft = TRUE,
  verbose = FALSE
)

Arguments

Details

Check out vignette for the examples of plotting RiboSeq metaplots over start and stop codons, so that you can verify visually whether this function detects correct shifts.

For how the Fourier transform works, see: isPeriodic
For how the changepoint analysis works, see: changePointAnalysis

NOTE: It will remove softclips from valid width, the CIGAR 3S30M is qwidth 33, but will remove 3S so final read width is 30 in ORFik. This is standard for ribo-seq.

Value

a data.table with lengths of footprints and their predicted coresponding offsets

References

https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-018-4912-6

See Also

Other pshifting: changePointAnalysis(), shiftFootprints(), shiftFootprintsByExperiment(), shiftPlots(), shifts_load(), shifts_save()

Examples

## Basic run
# Transcriptome annotation ->
gtf_file <- system.file("extdata/references/danio_rerio", "annotations.gtf", package = "ORFik")
# Ribo seq data ->
riboSeq_file <- system.file("extdata/Danio_rerio_sample", "ribo-seq.bam", package = "ORFik")
## Not run: 
footprints <- readBam(riboSeq_file)
## Using CDS start site as reference point:
detectRibosomeShifts(footprints, gtf_file)
## Using CDS start site and stop site as 2 reference points:
#detectRibosomeShifts(footprints, gtf_file, stop = TRUE)
## Debug and detailed information for accepted reads lengths and p-site:
detectRibosomeShifts(footprints, gtf_file, heatmap = TRUE, verbose = TRUE)
## Debug why read length 31 was not accepted or wrong p-site:
#detectRibosomeShifts(footprints, gtf_file, must.be.periodic = FALSE,
#              accepted.lengths = 31, heatmap = TRUE, verbose = TRUE)

## Subset bam file
param = ScanBamParam(flag = scanBamFlag(
                       isDuplicate = FALSE,
                       isSecondaryAlignment = FALSE))
footprints <- readBam(riboSeq_file, param = param)
detectRibosomeShifts(footprints, gtf_file, stop = TRUE)

## Without 5' Annotation
library(GenomicFeatures)

txdb <- loadTxdb(gtf_file)
tx <- exonsBy(txdb, by = "tx", use.names = TRUE)
tx <- extendLeaders(tx, 30)
## Now run function, without 5' and 3' UTRs
detectRibosomeShifts(footprints, txdb, start = TRUE, minFiveUTR = NULL,
                     minCDS = 150L, minThreeUTR = NULL, firstN = 150L,
                     tx = tx)


## End(Not run)

Description

Disengagement score is defined as

(RPFs over ORF)/(RPFs downstream to transcript end)

A pseudo-count of one is added to both the ORF and downstream sums.

Usage

disengagementScore(
  grl,
  RFP,
  GtfOrTx,
  RFP.sorted = FALSE,
  weight = 1L,
  overlapGrl = NULL
)

Arguments

Value

a named vector of numeric values of scores

References

doi: 10.1242/dev.098344

See Also

Other features: computeFeatures(), computeFeaturesCage(), countOverlapsW(), distToCds(), distToTSS(), entropy(), floss(), fpkm(), fpkm_calc(), fractionLength(), initiationScore(), insideOutsideORF(), isInFrame(), isOverlapping(), kozakSequenceScore(), orfScore(), rankOrder(), ribosomeReleaseScore(), ribosomeStallingScore(), startRegion(), startRegionCoverage(), stopRegion(), subsetCoverage(), translationalEff()

Examples

ORF <- GRanges(seqnames = "1",
               ranges = IRanges(start = c(1, 10, 20), end = c(5, 15, 25)),
               strand = "+")
grl <- GRangesList(tx1_1 = ORF)
tx <- GRangesList(tx1 = GRanges("1", IRanges(1, 50), "+"))
RFP <- GRanges("1", IRanges(c(1,10,20,30,40), width = 3), "+")
disengagementScore(grl, RFP, tx)

Description

Will calculate distance between each ORF end and begining of the corresponding cds (main ORF). Matching is done by transcript names. This is applicable practically to the upstream (fiveUTRs) ORFs only. The cds start site, will be presumed to be on + 1 of end of fiveUTRs.

Usage

distToCds(ORFs, fiveUTRs, cds = NULL)

Arguments

Value

an integer vector, +1 means one base upstream of cds, -1 means 2nd base in cds, 0 means orf stops at cds start.

References