Graph read simulator

This is a simple collection of scripts for simulating reads for a diploid genome by using vg and genome graphs

Running

graph_read_simulator simulate_reads -s 0.01 CHROMOSOME_HAPLOTYPE COVERAGE

Complex running (using GNU parallel, e.g. on a cluster)

Step 1: Make config files with chromosomes and haplotypes

chromosomes.txt (list of chromosome names, one on each line):

21
22

haplotypes.txt (list of chromosome names paired with haplotype numbers (either 0 or 1)):

21 0
21 1
22 0
22 1

Step 2: Init some config variables and run the prepare data script

This will create all the graphs, intervals etc needed for simulating reads

population_vcf="population.vcf.gz"
individual_vcf="individual.vcf.gz"
linear_ref_fasta="ref.fa"
chromosomes="21,22"
n_chromosomes=2
individual_ID="HG002"

scripts/graph_read_simulator_prepare_data $population_vcf $individual_vcf $linear_ref_fasta $chromosomes $n_chromosomes $individual_ID

Step 3: Simulate the reads

coverage=15
cat haplotypes.txt | parallel --line-buffer -j 2 "graph_read_simulator simulate_reads -s 0.01 {} $coverage" | graph_read_simulator assign_ids positions.tsv simulated_reads.fa

Note: The simulate_reads command can be run with different parameters, run graph_read_simulator simulate_reads to see available options.