outputVCF: {description: "Output VCF file. Add .gz to the file name to get compressed output. If omitted, use standard output.", category: "common"}
reference: {description: "Reference file. Provide this to detect alleles through re-alignment. If no index (.fai) exists, it will be created", category: "common"}
no_reference: {description: "Detect alleles without requiring a reference, at the expense of phasing quality (in particular for long reads)", category: "common"}
tag: {description: "Store phasing information with PS tag (standardized) or HP tag (used by GATK ReadBackedPhasing) (default: {description: PS)", category: "common"}
output_read_list: {description: "Write reads that have been used for phasing to FILE.", category: "advanced"}
algorithm: {description: "Phasing algorithm to use (default: {description: whatshap)", category: "advanced"}
merge_reads: {description: "Merge reads which are likely to come from the same haplotype (default: {description: do not merge reads)", category: "common"}
internal_downsampling: {description: "Coverage reduction parameter in the internal core phasing algorithm. Higher values increase runtime *exponentially* while possibly improving phasing quality marginally. Avoid using this in the normal case! (default: {description: 15)", category: "advanced"}
indels: {description: "Also phase indels (default: {description: do not phase indels)", category: "common"}
ignore_read_groups: {description: "Ignore read groups in BAM/CRAM header and assume all reads come from the same sample.", category: "advanced"}
sample: {description: "Name of a sample to phase. If not given, all samples in the input VCF are phased. Can be used multiple times.", category: "common"}
chromosome: {description: "Name of chromosome to phase. If not given, all chromosomes in the input VCF are phased. Can be used multiple times.", category: "common"}
error_rate: {description: "The probability that a nucleotide is wrong in read merging model (default: {description: 0.15).", category: "advanced"}
maximum_error_rate: {description: "The maximum error rate of any edge of the read merging graph before discarding it (default: {description: 0.25).", category: "advanced"}
threshold: {description: "The threshold of the ratio between the probabilities that a pair of reads come from the same haplotype and different haplotypes in the read merging model (default: {description: 1000000).", category: "advanced"}
negative_threshold: {description: "The threshold of the ratio between the probabilities that a pair of reads come from different haplotypes and the same haplotype in the read merging model (default: {description: 1000).", category: "advanced"}
full_genotyping: {description: "Completely re-genotype all variants based on read data, ignores all genotype data that might be present in the VCF (EXPERIMENTAL FEATURE).", category: "experimental"}
distrust_genotypes: {description: "Allow switching variants from hetero- to homozygous in an optimal solution (see documentation).", category: "advanced"}
include_homozygous: {description: "Also work on homozygous variants, which might be turned to heterozygous", category: "advanced"}
default_gq: {description: "Default genotype quality used as cost of changing a genotype when no genotype likelihoods are available (default 30)", category: "advanced"}
gl_regularize_r: {description: "Constant (float) to be used to regularize genotype likelihoods read from input VCF (default None).", category: "advanced"}
changed_genotype_list: {description: "Write list of changed genotypes to FILE.", category: "advanced"}
ped: {description: "Use pedigree information in PED file to improve phasing (switches to PedMEC algorithm). Columns 2, 3, 4 must refer to child, mother, and father sample names as used in the VCF and BAM/CRAM. Other columns are ignored.", category: "advanced"}
recombination_list: {description: "Write putative recombination events to FILE.", category: "advanced"}
recomb_rate: {description: "Recombination rate in cM/Mb (used with --ped). If given, a constant recombination rate is assumed (default: {description: 1.26cM/Mb).", category: "advanced"}
gen_map: {description: "File with genetic map (used with --ped) to be used instead of constant recombination rate, i.e. overrides option --recombrate.", category: "advanced"}
no_genetic_haplo_typing: {description: "Do not merge blocks that are not connected by reads (i.e. solely based on genotype status). Default: {description: when in --ped mode, merge all blocks that contain at least one homozygous genotype in at least one individual into one block.", category: "advanced"}
use_ped_samples: {description: "Only work on samples mentioned in the provided PED file.", category: "advanced"}
vcf: {description: "VCF or BCF file with variants to be phased (can be gzip-compressed)", category: "required"}
vcfIndex: {description: "Index for the VCF or BCF file with variants to be phased", category: "required"}
phaseInput: {description: "BAM, CRAM, VCF or BCF file(s) with phase information, either through sequencing reads (BAM, CRAM) or through phased blocks (VCF, BCF)", category: "required"}
memory: {description: "The amount of memory this job will use.", category: "advanced"}
timeMinutes: {description: "The maximum amount of time the job will run in minutes.", category: "advanced"}
dockerImage: {description: "The docker image used for this task. Changing this may result in errors which the developers may choose not to address.", category: "advanced"}
sample: "Name of the sample to process. If not given, use first sample found in VCF."
chr_lengths: "File with chromosome lengths (one line per chromosome, tab separated '<chr> <length>') needed to compute N50 values."
tsv: "Filename to write statistics to (tab-separated)."
only_sn_vs: "Only process SNVs and ignore all other variants."
block_list: "Filename to write list of all blocks to (one block per line)."
chromosome: "Name of chromosome to process. If not given, all chromosomes in the input VCF are considered."
vcf: "Phased VCF file"
memory: {description: "The amount of memory this job will use.", category: "advanced"}
timeMinutes: {description: "The maximum amount of time the job will run in minutes.", category: "advanced"}
dockerImage: {description: "The docker image used for this task. Changing this may result in errors which the developers may choose not to address.", category: "advanced"}
outputFile: "Output file. If omitted, use standard output."
reference: "Reference file. Provide this to detect alleles through re-alignment. If no index (.fai) exists, it will be created"
regions: "Specify region(s) of interest to limit the tagging to reads/variants overlapping those regions. You can specify a space-separated list of regions in the form of chrom:start-end, chrom (consider entire chromosome), or chrom:start (consider region from this start to end of chromosome)."
ignore_linked_read: "Ignore linkage information stored in BX tags of the reads."
linked_read_distance_cut_off: "Assume reads with identical BX tags belong to different read clouds if their distance is larger than LINKEDREADDISTANCE (default: 50000)."
ignore_read_groups: "Ignore read groups in BAM/CRAM header and assume all reads come from the same sample."
sample: "Name of a sample to phase. If not given, all samples in the input VCF are phased. Can be used multiple times."
output_haplo_tag_list: "Write assignments of read names to haplotypes (tab separated) to given output file. If filename ends in .gz, then output is gzipped."
tag_supplementary: "Also tag supplementary alignments. Supplementary alignments are assigned to the same haplotype the primary alignment has been assigned to (default: only tag primary alignments)."
vcf: "VCF file with phased variants (must be gzip-compressed and indexed)"
alignments: "File (BAM/CRAM) with read alignments to be tagged by haplotype"
memory: {description: "The amount of memory this job will use.", category: "advanced"}
timeMinutes: {description: "The maximum amount of time the job will run in minutes.", category: "advanced"}
dockerImage: {description: "The docker image used for this task. Changing this may result in errors which the developers may choose not to address.", category: "advanced"}
