snakemake.yml

# You may provide default values for environment variable. They will only be
# used if the variable is undefined in the current environment.
#default_env:
#    TMPDIR: /home/user/tmp

# You may also override values in the environment.
#override_env:
#    LANG:   C


# --- Environment variables substitution ---
#
# Environment variables may be referenced in the config via `$VARNAME` or
# `${VARNAME}`; to get a literal dollar sign use `$$`. Values in `deafult_env`
# `override_env` are expanded with the unmodified environment.
#
# Beware, if using environment variables in `dentist_config` the data type
# must be correct, e.g. you cannot pass `read-coverage` be environment
# variable because it will be a string but a float is expected. Still, if you
# specify values via `env` in the config then they will keep their data type,
# so the following will work:
#
#     override_env:
#         COV: 30.0
#     dentist_config:
#         __default__:
#             read-coverage: $COV


# Uncomment the follow workflow flags to alter the workflow

# Add the validation report and weak coverage mask to the list of default
# targets so they will be generated wether or not they are required to build
# the main target (gap-closed assembly)
#full_validation: true

# By default the closed gaps will be validated and only valid gaps will be
# included into the final gap-closed assembly. Give this flag to get the
# raw gap-closed assembly as final output.
#no_purge_output: true

# The workflow can be easily run inside a container (--use-singularity) to
# reduce the software installations. The default container is
# "docker://aludi/dentist:stable" and can be changed by the following line.
#dentist_container: "docker://aludi/dentist:edge"

# The workflow can be easily run inside using conda (--use-conda) to
# reduce manual software installations. The default Conda environment
# definition "envs/dentist_v2.yml" and can be changed by the following line.
#dentist_env: "envs/dentist_local.yml"

# Config file for dentist. Use this file to adjust parameters of DENTIST.
# You must set at least either `ploidy` and `read-coverage` or
# `max-coverage-reads` and `min-coverage-reads`.
dentist_config:         dentist.yml

inputs:
    # The reference assembly where gaps should be closed
    reference:          reference.fasta
    # The set of long reads used for gap closing
    reads:              reads.fasta
    # Type of reads. Use `PACBIO_SMRT` or `OXFORD_NANOPORE`. See README for
    # more details on the subject.
    reads_type:         PACBIO_SMRT

outputs:
    # The gap-closed reference assembly
    output_assembly:    gap-closed.fasta
    # JSON file with a validation report for each closed gap
    #validation_report:  $workdir/validation-report.json

reference_dbsplit:
    - -x1000
    - -a
    # Increase this value to reduce the workload of the cluster scheduler;
    # this increases the runtime and memory usage per alignment job
    - -s200
reads_dbsplit:
    - -x1000
    - -a
    # Increase this value to reduce the workload of the cluster scheduler;
    # this increases the runtime and memory usage per alignment job
    - -s200


# Directories for intermediate and log files.
workdir:            workdir
logdir:             logs

# Workflow files that are created inside of workdir; uncomment and adjust if
# you feel the need
#workflow:
#    self_mask:          dentist-self
#    reads_mask:         dentist-reads
#    pile_ups:           pile-ups.db
#    insertions_batch:   insertions/batch.{batch_id}.db
#    dentist_merge_config_file: dentist.merge.json
#    insertions:         insertions.db
#    closed_gaps_mask:   closed-gaps
#    weak_coverage_mask: dentist-weak-coverage

# Maximum number of threads per process if enough are available. Do not use
# too many threads (say <=32) as the speedup as usually sublinear in the
# number of threads. Very high numbers may cause segmentation faults.
threads_per_process:  8

# Number of threads to use for auxiliary commands such as daligner or daccord.
# Defaults to max(1, threads // 4)
#auxiliary_threads: 1

# Masks are propagated from the assembly to the reads and back in batches
# of read DB blocks. This makes cluster jobs run for at least a few minutes
# as each block usually just takes much less than a minute.
propagate_batch_size: 50

# Insertions are generated in batches of `batch_size`; use lower number to
# shorten the runtime of each job while increasing the number of jobs.
batch_size:         50

# Validation is split into this number of similar-sized parts of the
# gap-closed vs reads alignment. If the number of blocks is higher than the
# number of contigs in the gap-closed assembly then some validation runs will
# do nothing.
validation_blocks:  32