generate

generate is the entry point for building or importing polygenic score models. It holds:

• generate$sources$...(): declare the GWAS summary statistics to fit from
• generate$algorithms$...(): declare the PRS method to fit with
• generate$models(): cross-join sources with algorithms and execute them
• generate$from.pgs.catalog(): import published models from the PGS Catalog
• generate$from.pgs.file(): read models from local PGS scoring files

A typical generate workflow declares sources and algorithms separately, then crosses them:

models <- generate$models(
  sources    = generate$sources$opengwas("ieu-b-2"),
  algorithms = c(
    generate$algorithms$ClumpingThresholding(
      pval = c(5e-8, 1e-5), reference.panel = "EUR"
    ),
    generate$algorithms$LDpred2(
      reference.panel = "EUR", variant.space = "HapMap3+"
    )
  )
)

Published models bypass fitting entirely:

models <- generate$from.pgs.catalog("PGS002280")
models <- generate$from.pgs.file("scores/")