Example data and R scripts for PTL.PRS

This record provides a minimal, reproducible toy dataset and companion R scripts to exercise the PTL.PRS package end-to-end (training and evaluation), using both individual-level genotypes and summary-statistics–based pseudo-tests. What’s included Scripts example_script.RBasic workflow: trains with PTL_PRS_train and evaluates on individual-level test data. example_script_pseudotest_multiSeeds.RBatch workflow: runs PTL_PRS_run across multiple random seeds and evaluates using summary statistics (reports pseudo-R²). Genetic data (toy) Target panel: 1000 Genomes Phase 3 — South Asian (SAS). Chromosome: 22 only. Variant counts: 16,210 SNPs in the source PRS; 613,839 SNPs in SAS PLINK files (chr22 subset used here). PLINK bundles: 1kg_SAS_chr22 — main SAS PLINK set (chr22). 1kg_SAS_ref — reference subset (200 samples), PLINK-formatted. Phenotypes & summaries 601 total target samples 200 reference → PLINK: 1kg_SAS_ref 300 train/validation → phenotype file; derived summary stats: tr_val.summaries 101 test → randomly generated binary phenotype (0/1); phenotype file (sample indices from FAM): test_ped.txt Source PRS t2d_prscs_EUR.pgs (example score file used as the starting PRS)