config.yaml

# ============================================================================
# Config file for RNA-seq
#
# ==================[ Sections for the users ]================================
#
# One of input_directory, input_pattern and input_samples must be provided
# If input_directory provided, use it otherwise if input_pattern provided,
# use it, otherwise use input_samples.
# ============================================================================
sequana_wrappers: "v0.15.1"

input_directory:
input_readtag: _R[12]_
input_pattern: '*fastq.gz'

# See sequana_pipetools.readthedocs.io for details about these 2 options
# common prefixes are removed. addition prefixes may be removed here
#extra_prefixes_to_strip = []
# in special cases, sample names can be extracted with a pattern
#sample_pattern: '{sample}.fastq.gz'
apptainers:
  sequana_tools: "https://zenodo.org/record/7102074/files/sequana_tools_0.14.3.img"
  salmon: "https://zenodo.org/record/5708843/files/salmon_1.3.0.img"
  fastqc: "https://zenodo.org/record/7015004/files/fastqc_0.11.9-py3.img"
  fastp: "https://zenodo.org/record/7319782/files/fastp_0.23.2.img"
  igvtools: "https://zenodo.org/record/7022635/files/igvtools_2.12.0.img"
  graphviz: "https://zenodo.org/record/7928262/files/graphviz_7.0.5.img"

# =========================================== Sections for the users

#############################################################################
# Genome section:
#
# :Parameters:
#
# - aligner: either star or bowtie2. 
# - genome_directory: directory where all indexes are written.
# - rRNA_contaminant: path to an existing fasta file for ribosomal RNA (to be found in
#   genome_directory)
# - rRNA_feature: if rRNA_contaminant is not provided, ribosomal RNA will be extract
#     from GFF using this feature name. It must be found. 
general:
    aligner: bowtie2
    genome_directory:
    contaminant_file:
    rRNA_feature: rRNA
    custom_gff: ''


#################################################################
# FastQC section
#
# :Parameters:
#
# - options: string with any valid FastQC options
#
fastqc:
    skip_fastqc_raw: true
    options: --nogroup
    threads: 4
    resources:
        mem: 4G

#######################################################################
# Quality trimming and adapter removal
#
# for cutadapt, please fill the fwd and rev fields if required. It can be a
# string, or a filename. If a filename, it must be prefixed with "file:" to
# specify that it is a file and not a string. If the tool is cutadapt, the empty
# fwd and rev fields means that no adapters are to be used.
#
# :Parameters:
#
# - fwd: a string or file (prefixed with *file:*)
# - m: 20 means discard trimmed reads that are shorter than 20.
#         must be > 0
# - quality: 0 means no trimming, 30 means keep base with quality
#        above 30
# - mode: must be set to one of
#     - g for 5'
#     - a for 3'
#     - b for both 5'/3'
# - rev: a string or file (prefixed with *file:*)
# - tool: only cutadapt supported for now
# - threads: number of threads to use (atropos only)
# - options: See cutadapt documentation for details on
#            cutadapt.readthedocs.io. We change the default value
#            of -O to 6 (at least 6 bases are required to match before
#            trimming of an adapter)
#
# tool_choice__ = ["atropos", "cutadapt"]
#
# trim-n trims Ns at the end of the read
cutadapt:
    tool_choice: cutadapt
    fwd: ''
    rev: ''
    m: 20                       # {"strict_min": 0}
    mode: b                     # {"values": ["b","g","a"]}
    options: -O 6 --trim-n
    quality: 30                 # {"range": [0,40]}
    threads: 4


#############################################################################
# -Q should disable the quality filter
#
# Quality filtering only limits the N base number (-n, --n_base_limit) 
# meaning if 5 Ns are found, the read is discarded, 
# -q is the quality value se to Q15 to be qualified; If more than 40% of bases
# are unqualified, the read is discarded.
# You can also filter reads by average quality score using -e QUAL_score
#
# minimum length is set to 15 by default
#
# Adapter trimming is set by default. Can be disable with -A
# For adapters, this is automatic but you can be specific using 
# --adapter_sequence for read1, and --adapter_sequence_r2 for read2.
# The --cut_tail moves a sliding window from tail (3') to front, drop the bases 
# in the window if its mean quality is below cut_mean_quality, stop otherwise. 
# Use cut_tail_window_size to set the widnow size (default 4)), and 
# cut_tail_mean_quality to set the mean quality threshold (default 20)
# Other useful options: --disable_adapter_trimming and --disable_quality_filtering.
# or -n 5 (minimum number of Ns required to discard a read)
fastp:
    options: ' --cut_tail '
    minimum_length: 20
    adapters: ''
    quality: 15
    threads: 4
    disable_adapter_trimming: false
    disable_quality_filtering: false

#######################################################
# Quality trimming software choice
#
# software_choice__ = ["atropos", "cutadapt", "fastp"]
#
trimming:
    software_choice: fastp
    do: true

#############################################################################
# bowtie1_mapping_rna used to align reads against ribosomal RNA
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored
# - options: any options recognised by bowtie1 tool
# - threads: number of threads to be used
# - nreads: no need to analyse all data to estimate the ribosomal content. 
#   analyse 100,000 reads by default. Set to -1 to ignore and analyse all data
bowtie1_mapping_rna:
    # remove in v1.20 and set automatically to on/off if rRNA/fasta provided
    # do: true
    options: ''
    threads: 4
    nreads: 100000

#############################################################################
# star_mapping used to align reads against genome file
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored
# - options: any options recognised by rna-star tool. Set limitBAMsortRAM to 30G
# - threads: number of threads to be used
# - legacy: if set to True will use the old 2-pass version from STAR
#      used in this pipeline until v0.15.3. If you want to use the
#      2-pass mode available in star, you will need star 2.7 and above
# 
star_mapping:
    options: " --limitBAMsortRAM 30000000000 --outFilterMismatchNoverLmax 0.05 --seedSearchStartLmax 20 "
    legacy: True
    threads: 4
    resources:
      mem: 32G

##############################################################################
# STAR indexing section
#
# :Parameters:
#
# - options: string with any valid STAR options
star_index:
    options:
    threads: 4
    resources:
      mem: 4G
#############################################################################
# bowtie1_mapping_ref used to align reads against genome file
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored
# - options: any options recognised by bowtie1 tool
# - threads: number of threads to be used
#
bowtie1_mapping_ref:
    options: --chunkmbs 400 -m 1
    threads: 4

#############################################################################
# bowtie2_mapping used to align reads against genome file
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored
# - options: any options recognised by bowtie2 tool
# - threads: number of threads to be used
#
bowtie2_mapping:
    #options: "--dovetail --no-mixed --no-discordant " for paired-end data
    options: ''
    threads: 4
    genome_size_larger_than_4gb: false
    resources:
      mem: 20G

bowtie2_index:
    options: ''
    threads: 4
    resources:
      mem: 20G

salmon_index:
    threads: 2
    options:
    resources:
        mem: 4G

salmon_mapping:
    options: -l A
    threads: 4
    resources:
        mem: 4G

#############################################################################
# feature_counts used to count reads against features
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored
# - options: any options recognised by feature_counts tool except -s
# - threads: number of threads to be used
# - strandness: (optional) you should provide the strand parameters, given
#      from the experimental design. If not provided, we will guess it (see
#      tolerance parameter here below)
# - tolerance: if strandness is not provided, we will guess it from
#     the data. The metric used is between 0 and 1. It is a ratio between 
#     strand + and -. If below tolerance, the strand is reversely stranded. If
#     above 1-tolerance, it is (forward) stranded. If around 0.5 +- tolerance,
#     it is unstranded. Otherwise, it means our guess would not be very
#     reliable. A warning will be raised. Note also that if there is no
#     consensus across samples, a warning/error may also be raised. tolerance
#     is therefore in the range [0-0.25]
# - feature: this is equivalent to the -t option to specify the feature type in GTF
#     annotation. For example gene, exon (default). 
# - attribute: this is the -g option to specify the attribute type in GTF annoation. 
#   (gene_id) by default. 
# - extra_attributes: any other 
#
feature_counts:
    do: true
    options: ''      ## if exon/CDS is used, put -O option
    strandness: ''   # set to 0, 1, 2 to force te type of strandness
    threads: 1       # 
    tolerance: 0.15  # use to figure out the strandness. no need to change
    feature: gene    # could be exon, mRNA, etc
    attribute: ID    # could be ID, gene_id, etc
    extra_attributes:    # by default, stores only the main attribute, but could add more

#############################################################################
# bamCoverage write file in bigwig format from BAM files.
# This tool takes an alignment of reads or fragments as input (BAM file) and
# generates a coverage track (bigWig or bedGraph) as output. The coverage is
# calculated as the number of reads per bin, where bins are short consecutive
# counting windows of a defined size. It is possible to extended the length of
# the reads to better reflect the actual fragment length. *bamCoverage* offers
# normalization by scaling factor, Reads Per Kilobase per Million mapped reads
# (RPKM), and 1x depth (reads per genome coverage, RPGC).
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored
# - binSize: Size of the bins, in bases, for the output of the
#            bigwig/bedgraph file. (default: 50)
# - genomeSize: Report read coverage normalized to 1x sequencing depth
#                        (also known as Reads Per Genomic Content (RPGC)).
#                        Sequencing depth is defined as: (total number of
#                        mapped reads * fragment length) / effective genome
#                        size. The scaling factor used is the inverse of the
#                        sequencing depth computed for the sample to match the
#                        1x coverage. To use this option, the effective genome
#                        size has to be indicated after the option. The
#                        effective genome size is the portion of the genome
#                        that is mappable.
# - extendReads: This parameter allows the extension of reads to
#                fragment size.
# - minFragmentLength: The minimum fragment length needed for read/pair
#                      inclusion. Note that a value other than 0 will exclude
#                      all single-end reads.
# - maxFragmentLength: The maximum fragment length needed for read/pair
#                      inclusion. A value of 0 disables filtering and is
#                      needed for including single-end and orphan reads.
# - threads: number of threads to be used
bam_coverage:
    do: false
    options:
    binSize: 10
    genomeSize: 2150570000  ##mm10
    extendReads: 65
    minFragmentLength: 0 #Note that a value other than 0 will exclude all single-end reads.
    maxFragmentLength: 0 #A value of 0 disables filtering and is needed for including single-end and orphan reads.
    threads: 4
    resources:
        mem: 20G


###########################################################################
# Creates a tdf files using igvtools
#
# :Parameters:
#
# - chromSize: path to index of reference genome obtain by samtools faidx
igvtools:
    do: false
    # can be a link to the fasta file or an existing chrom.sizes file
    # If none provided, will use the input fasta file
    chrom_sizes_file: ''


#############################################################################
# mark_duplicates (picard-tools) allows to mark PCR duplicate in BAM files
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored. Mandatory for RNA-SeQC tool.
# - remove: If true do not write duplicates to the output file instead of writing them with
#            appropriate flags set.  Default value: false. This option can be set to 'null' to clear
#            the default value. Possible values: {true, false}
# - tmpdir: write tempory file on this directory (default TMP_DIR=/tmp/, but could be "TMP_DIR=/local/scratch/")
#
mark_duplicates:
    do: false
    remove: false ## may be True
    tmpdir: ./tmp/
    threads: 4
    resources:
      mem: 34G

add_read_group:
    options: 

#############################################################################
# RNA-SeQC allows to compute a series of quality control metrics for RNA-seq data
#
# :Parameters:
#
# - do: if unchecked, this rule is ignored
# - ref: Reference Genome in fasta format
# - gtf: GTF File defining transcripts (must end in '.gtf')
#        You can use the 'sequana gff-to-gtf input.gff' command
# - options: any options recognised by RNA-seQC tool
rnaseqc:
    do: false
    gtf_file:
    options: --coverage


# if be_file not provided, try to create one on the fly
# needs mark_duplicates
rseqc:
    do: false
    bed_file:


#############################################################################
#   MultiQC aggregates results from bioinformatics analyses across many
#   samples into a single report.
#
# :Parameters:
#
# - options: any options recognised by multiqc
# - output-directory: Create report in the specified output directory
# - config_file: by default, we use sequana RNA-seq multiqc_config file. 
#       If you want your own multiqc, fill this entry
multiqc:
    options: -p -f -x *_init_*
    modules: ''
    input_directory: .
    config_file: multiqc_config.yaml