The complete scripts are available below. Website builds display code only; they never run these scripts or download biological inputs.
Local end-to-end workflow
View end_to_end_phenotype.R or read the complete script on GitHub.
With gsim and qgg already installed, run from the repository root:
The script creates a small synthetic phased VCF fixture locally, generates a reusable founder population, transmits chromosomes through a pedigree, writes HAP and BED, prepares a qgg Glist, and simulates phenotypes. It compares default, causal-probability, variance-weight, and combined configurations. Its synthetic LD scores illustrate alignment and parameterization, not real reference LD.
qgg is optional for gsim generally but required for this Glist workflow. The phenotype stage uses bounded BED decoding and native accumulation; HAP phenotype input is not supported. The phenotype contract details memory bounds. Compatibility with gsuite-generated Glist objects is not established.
1000 Genomes chromosome 22
View 1000G_chr22.R or read the complete script on GitHub.
This external-data workflow requires internet access when the
official IGSR GRCh37 VCF and HapMap map are absent. The VCF download is
approximately 196 MB. The script documents GSIM_1000G_DIR
for local input reuse. It imports a selected region and named samples,
creates unrelated founders, and simulates pedigree descendants. This is
a real 1000G reference workflow, unlike the synthetic local fixture
above. No data are bundled with the website.
The following command is an external-input workflow, not a website build step:
source("inst/examples/1000G_chr22.R")See Getting started for the stage-by-stage workflow sketches and Validation and performance for the retained scientific evidence, benchmark environments, and uncertainty. External HAPNEST comparison remains deferred; pedigree meiosis remains single-threaded.
Controlled gene sets
Run source("inst/examples/gene_sets.R") from the
repository root with gsim installed. The example obtains causal SNPs
from a small existing simulation, uses an explicitly synthetic mapping,
and prints exact set composition and shared genes. Sets are deliberately
conditional on causal truth.
View gene_sets.R. See the authoritative contract for input, sampling and truth semantics and instructions for substituting a real mapping.