##############################
# Variables you need to change
##############################

samples: "/n/holyscratch01/davis_lab/dwhite/snpArcher_old/config/samples_test10.csv" # path to the sample metadata CSV
resource_config: "config/resources.yaml" # path to resources yaml config
final_prefix: "test10" # prefix for final output files
intervals: True #Set to True if you want to perform variant calling using interval approach.
sentieon: False #set to True if you want to use sentieon, False if you want GATK
sentieon_lic: "" #set to path of sentieon license
remote_reads: False # Set True if reads are in a Google Bucket seperate from --default-remote-prefix.
remote_reads_prefix: "" # set to google bucket prefix where reads live
bigtmp: "" #Set to a path with lots of free space to use for commands that require large amounts of temp space; defaults to system tmpdir if empty
cov_filter: True #set to True if you want to include coverage thresholds in the callable sites bed file (default uses mappability only)
generate_trackhub: True #Set to true if you want to generate a Genome Browser Trackhub. Dependent on postprocessing module.
trackhub_email: ""
##############################
# Variables you *might* need to change
##############################

# Interval approach options, only applicable if intervals is True
minNmer: 500 # the minimum Nmer used to split up the genome; e.g. a value of 200 means only Nmers 200 or greater are used to define the boundaries of intervals. The minimum is 50.
num_gvcf_intervals: 200 # The maximum number of intervals to create for GVCF generation. Note: the actual number of intervals may be less than the specified value if the reference genome has very few gaps.
db_scatter_factor: 0.15 # Scatter factor for calculating number of intervals to create for genomics db generation. (scatter_factor * num_samples * num_gvcf_intervals) gives us number of db intervals to create. Reccomend <1

## Coverage options ##
## default pipeline is optimized for low coverage data - if using high coverage data (> 10x), uncomment high coverage options and comment out low coverage options

# low coverage options (< 10x)
minP: 1
minD: 1

# high coverage options (> 10x)
#minP: 2
#minD: 4

het_prior: 0.005 #prior probabilty of heterozygous site; changes likelihood of a site being called non-ref, but not genotype likelihoods

## callable sites bed file options ##
mappability_min: 1 #regions of the genome with mappability less than this will be removed from callable sites bed file
mappability_k: 150 #the kmer used to compute mappability with genmap; you should not need to change this except in special cases

#this ignores small regions of abberatant coverage/mappability as often these are just below the threshold
#to do strict filtering, set to 0

mappability_merge: 100 #merge passing mappability regions separated by this or fewer bp into a single region
cov_merge: 100 #merge passing coverage regions separate by this or fewer bp into a single region

## QC options ##
nClusters: 3
GoogleAPIKey:
min_depth: 2

## Filtering options ##

contig_size: 10000 # snps on contigs this size or smaller will be filtered from the final clean vcfs. Set to 0 to disable.
maf: 0.01 #snps with MAF below this value will be filtered from the final clean vcfs. Set to 0 to disable.
missingness: 0.75 #snps with missingness greater than this value will be filtered from the final clean vcfs. Set to 1 to disable.
scaffolds_to_exclude: "mtDNA,Y" #comma separated, no spaces list of scaffolds to exclude from final clean vcfs. Set to blank to disable.

########################################
## coverage thresholds ##
########################################

## If cov_filter is True, use these parameters to control how coverage filtering is done
## Three options are provided for coverage-based filtering. The default option is to just filter
## regions of the genome with mean coverage below a minimal threshold (default = 1), with a very large upper limit
## To use this option, set the variables below to the lower absolute mean coverage limit and upper absolute mean coverage limit, 
## and make sure all other coverage variables are empty

cov_threshold_lower: 1
cov_threshold_upper: 10000
 
## Alternatively, filtering can be done based on standard deviations 
## (assumes a Poisson distribution, so stdev_cov equals the square root of the mean coverage),
## where regions of the genome with mean coverage < X standard deviations or > X standard deviations are removed.
## To use this option, set the variables below to the desired X
## and make sure all other coverage variables are empty

cov_threshold_stdev: 

## Finally, filtering can be done based on absolute scaling of the mean, 
## where regions of the genome with mean coverage < (global mean coverge / X) or > (global mean coverge * X) are removed.
## To use this option, set the variable below to the desired X
## and make sure all other coverage variables are empty

cov_threshold_rel: