pipeline.yml

##########################################################
##########################################################
##########################################################
## Default configuration file for intervals pipeline
##########################################################

# directory where input data are located, specified as:
# "." for the current directory
# alternative a path may be specified. The convention is to
# use 'data.dir'
# The default is '.'
input: .

# location of indexed genome 
genome: hg38_noalt

# location of indexed genome 
genome_dir: /shared/sudlab1/General/mirror/genomes/index






# scratchdir for data not to be backed up
scratchdir: /tmp

# a version string
version: ''

# directory for publishing results on the web
web_dir: ../web

# directory in which to save files for export
exportdir: export

# wether or not to preprocess the intervals
# preprocessing involves:
# * sorting the intervals
# * merging overlapping intervals
# * renaming intervals
# * removing strand and score information
preprocess: 0


##methods for motif finding
# currently supported are 
#
# MEME
# discMEME - discriminative MEME
# randDREME - DREME with ranomdised background
# discDREME - discriminative DREME
# descriminative methods require a design file (design.tsv) matching test 
# sets to negatives
# The design file must have a tab separated header
methods: meme, randDREME

###############################################################


database:
    url: sqlite:///.csvdb


    ################################################################
    ################################################################
    ################################################################
    ## track information
    ################################################################
bams:

offsets:


    ################################################################
    ################################################################
    ################################################################
    ## Location of annotation database
    ################################################################
annotations:
    database: /shared/sudlab1/General/annotations/hg38_noalt_ensembl85/csvdb

    # directory with annotation information
    dir: /shared/sudlab1/General/annotations/hg38_noalt_ensembl85

    #######################################################
    #######################################################
    #######################################################
    ## Parameters for computing peak shapes
    #######################################################
peakshape:
    # window width for computing peakshape
    window_size: 1000

    # bin size for computing peak shapes
    bin_size: 10

    #######################################################
    #######################################################
    #######################################################
    ## options related to motif search
    #######################################################
    #
    #Each tool gets a seperate section with common options
    #for selecting the sequence used to find motifs.
    #
    # these are:
    #
    # max_size:
    # maximum of the sum of lengths of sequences used for motif
    # searches. 
    #
    # halfwidth:
    # half width of region around peak to use for motif discovery
    # leave unset to use whole interval
    #
    # masker:
    # masker to use for meme, valid options are 'dustmasker' and 'repeatmasker'
    # (the latter requires an appropriately annotated genome.fasta file)
    #
    # score:
    # order of peaks to use, one of "peakval", "score" or "random".
    #
    #    score: use the score field of the bed files, higher is better
    #    peakval: use a peakval score calculated from from associated bamfiles
    #    random: randomise the order of intervals
    #
    # if an option is specificed but data (or bam files) absent, a score of
    # 1 is set for all peaks.
    #
    # Then a set of options to decide how many peaks to use:
    # Set either proportation (0 < proportion <= 1) and min sequences or 
    # num_sequences.
    #
    # the maximum of proportion and min_sequences
    # will be used to detect motifs
    # 

    #########################################################
    # MEME
    #########################################################

meme:
    ### Sequence file generation
    ## Selecting sequences
    #ranking (score, peakval or random)
    score: random

    # number of sequences to use
    num_sequences: ''
    # or
    proportion: 1
    min_sequences: 100

    ## Sequence manipulations
    halfwidth: ''
    masker: ''

    # meme model to use
    model: anr

    # maximum number of characters for motif discovery
    max_size: 200000000

    # number of motifs to find with meme
    nmotifs: 40

    # Threads to use
    # If the threads used are >1, additional options can be specified
    # In Iceberg, the following specifications for job and cluster parallel 
    # environments should be specified:
    # job_options=-l excl=false -w n
    # cluster_parallel_environment=openmpi-ib
    threads: 25

    # Options used if threads >1
    job_options: -l excl=false -w n
    cluster_parallel_environment: openmpi-ib



    # meme_options
    options: -minw 5 -maxw 30


psp:
    options: ''

    ############################################################
    # DREME
    ############################################################

dreme:
    ### Sequence file generation

    ## Selecting sequences
    #ranking (score, peakval or random)
    score: peakval

    # number of sequences to use
    num_sequences: ''
    # or
    proportion: 1
    min_sequences: 100

    ## Sequence manipulations
    max_size: ''
    halfwidth: ''
    masker: dustmasker

    # dreme_options
    options: -minw 5 -maxw 30

    ############################################################
    # MEME-ChIP
    ############################################################
memechip:
    ### Sequence file generation

    ## Selecting sequences
    #ranking (score, peakval or random)
    score: peakval

    # number of sequences to use
    num_sequences: ''
    # or
    proportion: 1
    min_sequences: 100

    ## Sequence manipulations
    max_size: ''
    halfwidth: 100
    masker: dustmasker

    #number of sequences to pass to MEME
    nmeme: 100


    # Threads to use
    # If the threads used are >1, additional options can be specified
    # In Iceberg, the following specifications for job and cluster parallel
    # environments should be specified:
    # job_options=-l excl=false -w n
    # cluster_parallel_environment=openmpi-ib
    threads: 1

    # Options used if threads >1
    job_options: -l excl=false -w n
    cluster_parallel_environment: openmpi-ib


    options: -meme-minw 5 -meme-maxw 30 -meme-nmotifs=3

    #############################################################
    # Biprospector
    #############################################################

bioprospector:
    #bioprospector options
    options: -W 6 -w 6 -g 3 -G 3

    ### Sequence file generation
    ## Selecting sequences
    #ranking (score, peakval or random)
    score: peakval

    # number of sequences to use
    num_sequences: ''
    # or
    proportion: 0.10
    min_sequences: 100

    ## Sequence manipulations
    max_size: 100000
    halfwidth: 100
    masker: dustmasker



glam2:
    # options for glam2: 100000 iterations (slow), motifs from 6 to 30, start at 20
    options: -r 5 -n 100000 -a 6 -b 30 -w 20
    # number of results to return for glam2scan
    scan_results: 20000

mast:
    # evalue threshold for mast - set to large value
    # to collect all results
    evalue: 10000000

    # options for mast
    options: ''



    ##########################################################
    ##########################################################
    ##########################################################
    # options for tomtom
    ##########################################################
tomtom:
    # maximum pvalue to filter MEME output using TOMTOM
    filter_pvalue: 1e-05

    # master motif to select motifs from MEME output using TOMTOM
    databases: [/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/JASPAR/JASPAR_CORE_2016.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/JASPAR/JASPAR_CORE_REDUNDANT_2016.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/JASPAR/JASPAR_CORE_2016_vertebrates.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/JASPAR/JASPAR_CORE_REDUNDANT_2016_vertebrates.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/HUMAN/HOCOMOCOv10_HUMAN_mono_meme_format.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/CIS-BP/Homo_sapiens.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/EUKARYOTE/wei2010_human_mws.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/EUKARYOTE/wei2010_mouse_mws.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/EUKARYOTE/wei2010_mouse_pbm.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/MOUSE/HOCOMOCOv10_MOUSE_mono_meme_format.meme,/shared/sudlab1/General/apps/bio/meme_4.11.2/db/motif_databases/CIS-BP/Mus_musculus.meme]

    # options for tomtom - defaults taken from tomtom web site
    options: -min-overlap 5 -dist pearson -evalue -thresh 10



    ################################################################
    ################################################################
    ################################################################
    ## gat options
    ################################################################
gat:

    # number of samples
    num_samples: 10000

    # mapability track to use
    mapability: 100

    # fdr to filter
    fdr: 0.05

    # minimum expected overlap for results to be reported
    min_expected: 1000

    # workspace to use for gat. If empty, the default
    # is to use the ungapped genome as defined in the
    # annotations pipeline
    workspace: ''

    ##############################################
    ##############################################
    ## read processing - filtering options
    ##############################################
filtering:

    # minimum mapping quality
    quality: 10

    # whether or not to dedup
    dedup: 1

    # method to remove duplicatesu
    dedup_method: picard

    # insert-size
    min_insert_size: 0

    # maximum insert size
    max_insert_size: 500

    ################################################################
geneset:
    # geneset to use to compute binding profiles for - see
    # pipeline_annotations ini file for definitions.
    binding: annotations_interface_geneset_all_gtf

    ################################################################
report:
    # number of threads to use to build the documentation
    threads: 10

    # directory for html documentation
    html: report/html

    # directory for doctrees
    doctrees: report/doctrees

    # prefix under which to publish report
    prefix: default