Configuration

Selscape uses multiple YAML configuration files to control different aspects of the analysis:

File	Purpose
config/main.yaml	Core settings (species, populations, data paths)
config/selscan.yaml	selscan parameters
config/betascan.yaml	BetaScan parameters
config/scikit-allel.yaml	scikit-allel parameters
config/dadi-cli.yaml	dadi-cli parameters
config/slurm/config.yaml	HPC cluster settings (optional)

main.yaml

Example

# Species identification
species: "Human"
tax_id: 9606
ref_genome: "hg38"
ploidy: 2

# Populations to analyze
populations:
  - YRI
  - CHS

# VCF file configuration
data_folder: "examples/data/Human/raw"
vcf_prefix: "full_chr"
vcf_suffix: ".vcf.gz"
# Files should be named: {vcf_prefix}{chromosome}{vcf_suffix}
# Example: full_chr21.vcf.gz

metadata: "examples/data/Human/metadata/example_metadata.txt"

# Chromosomes to analyze
chromosomes:
  - 20
  - 21

# Ancestral alleles for polarization (optional but recommended)
anc_alleles:
  path: "examples/data/Human/ancestral_alleles/homo_sapiens_ancestor_GRCh38"
  prefix: "homo_sapiens_ancestor_chr"
  # Files: {prefix}{chromosome}.bed.gz

# Annotation files
genome_annotation: "examples/data/Human/annotation/Human.gtf.gz"
gene2go: "examples/data/Human/annotation/gene2go.gz"

# Quality control
hwe_pvalue: 0.001  # Hardy-Weinberg equilibrium threshold

# Repeat regions (optional)
rmsk: "examples/data/Human/repeats/hg38.rmsk.autosomes.bed"
seg_dup: "examples/data/Human/repeats/hg38.seg.dups.autosomes.bed"
sim_rep: "examples/data/Human/repeats/hg38.simple.repeats.autosomes.bed"

# Method configuration files
selscan_config: "config/selscan.yaml"
betascan_config: "config/betascan.yaml"
dadi_config: "config/dadi-cli.yaml"
scikit_allel_config: "config/scikit-allel.yaml"

Parameters

Parameter	Type	Description	Required
species	string	Species name	Y
tax_id	integer	NCBI taxonomy ID	Y
ref_genome	string	Reference genome build	Y
anc_alleles.path	string	Base folder for ancestral allele information	N
anc_alleles.prefix	string	Ancestral allele information filename prefix	N
genome_annotation	string	Path to genome annotation GTF file	Y
gene2go	string	Path to gene2go mapping file	Y
hwe_pvalue	float	Hardy-Weinberg equilibrium p-value threshold	Y
hwe_pvalue	float	Hardy-Weinberg equilibrium p-value threshold	Y
rmsk	string	Path to repeat masker BED file	N
seg_dup	string	Path to segmental duplications BED file	N
sim_rep	string	Path to simple repeats BED file	N
ploidy	integer	Organism ploidy	Y
populations	list	List of population IDs to analyze	Y
data_folder	string	Base folder for input VCF files	Y
vcf_prefix	string	VCF filename prefix	Y
vcf_suffix	string	VCF filename suffix	Y
metadata	string	Path to sample metadata file	Y
chromosomes	list	List of chromosomes to analyze	Y
betascan_config	string	Path to BetaScan configuration file	Y
selscan_config	string	Path to selscan configuration file	Y
dadi_config	string	Path to dadi-cli configuration file	Y
scikit_allel_config	string	Path to scikit-allel configuration file	Y

betascan.yaml

Example

# Data type
unfolded: True  # Use polarized data (requires ancestral alleles)

# Core allele frequency parameter
core_frq: 0.15

# Allele frequency filters
min_af: 0.05  # Minimum allele frequency
max_af: 0.95  # Maximum allele frequency

# Statistical thresholds
top_proportion: 0.0005  # Top proportion for outlier identification

# Manhattan plot settings
manhattan_plot_width: 640
manhattan_plot_height: 240
manhattan_plot_color1: "#56B4E9"
manhattan_plot_color2: "#F0E442"

Parameters

Parameter	Type	Description	Required
unfolded	boolean	Use polarized/unfolded allele frequency spectrum	Y
core_frq	float	Core allele frequency parameter	Y
min_af	float	Minimum allele frequency	Y
max_af	float	Maximum allele frequency	Y
top_proportion	float	Top proportion for outlier identification	Y

selscan.yaml

Example

# Data type
unphased: true  # Set to false if you have phased data

# Within-population statistics
wp_stats:
  - ihs   
  - nsl    

# Cross-population statistics  
xp_stats:
  - xpehh  
  - xpnsl

# Minor allele frequency threshold
maf: 0.05

# Statistical thresholds
top_proportion: 0.0005

# Manhattan plot settings
manhattan_plot_width: 640
manhattan_plot_height: 240
manhattan_plot_color1: "#56B4E9"
manhattan_plot_color2: "#F0E442"

Parameters

Parameter	Type	Description	Required
unphased	boolean	Use unphased (true) or phased (false) data	Y
wp_stats	list	Within-population statistics to compute	Y
xp_stats	list	Cross-population statistics to compute	Y
maf	float	Minor allele frequency threshold	Y
top_proportion	float	Top proportion for outlier identification	Y

scikit-allel.yaml

Example

# Within-population statistics
wp_stats:
  - windowed_tajima_d   
  - moving_tajima_d  

# Windowed approach (base pair windows)
windowed_window_sizes: [100_000]
windowed_step_size_ratios: [1]  # 1 = non-overlapping windows

# Moving approach (SNP windows)
moving_window_sizes: [100]
moving_step_size_ratios: [1]

# Cross-population statistics 
xp_stats:
  - delta_moving_tajima_d

# XP moving approach (SNP windows)
delta_moving_window_sizes: [100]
delta_moving_step_size_ratios: [1]

# Statistical thresholds
top_proportion: 0.05

# Manhattan plot settings
manhattan_plot_width: 640
manhattan_plot_height: 240
manhattan_plot_color1: "#56B4E9"
manhattan_plot_color2: "#F0E442"

Parameters

Parameter	Type	Description	Required
wp_stats	list	Within-population statistics to compute	Y
xp_stats	list	Cross-population statistics to compute	Y
wtjd_window_sizes	list	Window sizes in base pairs for windowed tajima's d (wtjd) statistic	Y
wtjd_step_size_ratios	list	Step size as fraction of window in base pair for wtjd statistic	Y
mtjd_window_sizes	list	Window sizes in SNPs for moving tajima's d (mtjd) statistic	Y
mtjd_step_size_ratios	list	Step size as fraction of window in SNPs for mtjd statistic	Y
dtjd_moving_window_sizes	list	Window sizes in SNPs for xp moving delta tajima's d (dtjd) statistic	Y
dtjd_moving_step_size_ratios	list	Step size as fraction of window in SNPs for cross population dtjd statistic	Y
top_proportion	float	Top proportion for outlier identification	Y

dadi-cli.yaml

Example

# Data type
unfolded: True

# Demographic models
demog_1d: two_epoch
demog_1d_p0: "0.5 5 0.5"
demog_1d_ub: "10 10 1"
demog_1d_lb: "10e-5 10e-5 0"

# DFE models
dfe_1d: lognormal
dfe_1d_p0: "5 5 0.5"
dfe_1d_ub: "100 1000 1"
dfe_1d_lb: "10e-5 10e-5 0"

# Grid sizes
dm_grid_size: "300 400 500"
dfe_grid_size: "300 400 500"

# Gamma points for generating caches
gamma_pts: 100

# Ratio of non-synonymous/synonymous mutation rates
ratio: 2.31

# Number of optimization runs
optimizations: 50

# Bootstrap parameters
bootstrap_replicates: 100
chunk_size: 1000000

# Frequency spectrum flags
mask_singletons: false

Parameters

Parameter	Type	Description	Required
unfolded	boolean	Use polarized/unfolded allele frequency spectrum	Y
demog_1d	string	1D demographic model name	Y
demog_1d_p0	string	Initial parameters for 1D demographic model	Y
demog_1d_ub	string	Upper bounds for 1D demographic model parameters	Y
demog_1d_lb	string	Lower bounds for 1D demographic model parameters	Y
dfe_1d	string	1D DFE model name	Y
dfe_1d_p0	string	Initial parameters for 1D DFE model	Y
dfe_1d_ub	string	Upper bounds for 1D DFE model parameters	Y
dfe_1d_lb	string	Lower bounds for 1D DFE model parameters	Y
dm_grid_size	string	Grid sizes for demographic inference	Y
dfe_grid_size	string	Grid for DFE inference	Y
gamma_pts	integer	Number of gamma points for cache generation	Y
ratio	float	Ratio of non-synonymous to synonymous mutation rates	Y
optimizations	integer	Number of optimization runs	Y
bootstrap_replicates	integer	Number of bootstrap replicates	Y
chunk_size	integer	Chunk size for bootstrapping	Y
mask_singletons	boolean	Whether to mask singleton variants	Y

Important Notes

1. Ancestral Alleles: If anc_alleles is not provided, polarization-dependent analyses (selscan, betascan with unfolded=True, dadi with unfolded=True) will not run.

2. VCF File Naming: VCF files must follow the pattern: {data_folder}/{vcf_prefix}{chromosome}{vcf_suffix}

   • Example with vcf_prefix: "full_chr": examples/data/Human/raw/full_chr21.vcf.gz
   • Example with vcf_prefix: "chr": resources/data/chr21.vcf.gz
   • Example with vcf_prefix: "": resources/data/21.vcf.gz

3. Metadata Format: Tab-separated file with columns: Sample and Population

4. Repeat Masking: Optional but recommended for balancing selection analysis. Set to empty string or null to disable.

5. Step Size Ratios: Values from 0 to 1 represent the fraction of window size to step.

   • Example: window size = 100, step size ratio = 0.1 → step size = 10

Please check Snakemake Configuration documentation for additional information.