{ "samples": 10665, "populations": 319, "variants": 302955, "expected_pairs": 50721, "samples_by_panel": { "raw": 9331, "AADR": 179, "Behar2010": 19, "tajik_xinjiang": 64, "lazaridis2016_HO": 570, "behar_full": 248, "manchu_korean": 34, "med_new": 67, "swchina": 153 }, "excluded_singleton_records": 1, "variant_counts": { "retained_direct": 302955, "absent_in_AADR": 16197 }, "allele_order_reversals": 212559, "source_correction": "Applied 2026-09-05 Tuscan ledger; Buryat and Kalmyk split from Mongol; other reviewed pools retained.", "duplicate_policy": "One representative per accepted genotype-screen component; five conflicting Papuan representations and 64 full-panel missing records excluded; 18 Ignore records excluded before screening.", "method": "Hudson ratio-of-averages recomputed from harmonized genotypes, with inferred single-allele adjustment and same-mask diploid sensitivity; 5000-SNP block jackknife. No imputation.", "zero_heterozygosity_records": 214, "small_sample_populations": {}, "minimum_population_n": 5, "excluded_below5_populations": 26, "excluded_below5_individuals": 66, "filtering_note": "N>=5; India source-label exclusions retained. Two AADR Turkish records restored and source Turkish additions screened against every existing individual. Same SNP universe and per-population coverage mask.", "coverage_rule": "For each pair, retain a SNP only when at least 80% of retained individuals AND at least 5 individuals have a genotype in EACH population. Count individuals, not allele copies. Apply the identical mask to adjusted and diploid estimates.", "coverage_minimum_fraction": 0.8, "coverage_minimum_people": 5, "minimum_pair_qualifying_snps_for_display": 50000, "population_revision": "Balkan pool split by verified source labels; Croatian cohorts combined after genotype checks; English split and HGDP-only Mongol retained.", "india_revision_excluded_individuals": 51, "genotype_prefix": "/media/davide/5f3be2d1-52b6-4624-8349-65bed45c5868/newfiles/final_raw_fst_grch37/data/balkan_split_20260908/atlas_final_20260908", "release_date": "2026-09-08", "reconciliation_summary": { "candidate_records": 1525, "identity_candidate_pairs": 328, "confirmed_duplicate_edges": 299, "individual_actions": { "excluded_duplicate_existing": 205, "ready_for_inclusion": 1136, "excluded_insufficient_called_SNPs": 37, "hold_ambiguous_label": 12, "excluded_duplicate_source": 94, "below_N5_after_deduplication": 13, "insufficient_population_coverage": 28 }, "ready_individuals": 1136, "ready_population_groups": 76, "ready_new_population_groups": 64, "existing_atlas_individuals": 9612, "live_website_changed": false, "scope": "Candidate cohorts from the 2026-09-08 reviewed local-source inventory.", "rules": { "identity": "At least 5000 shared SNPs and >=99% diploid genotype concordance; pseudohaploid comparisons require >=5000 jointly homozygous SNPs and >=99.5% agreement. Same named samples with platform discordance additionally require >=50000 shared SNPs, >=98% total concordance, >=99.95% homozygous concordance and KING >=0.45. All identity decisions require genotype confirmation.", "representative": "Preserve existing curated atlas individual; otherwise diploid, then greatest call count.", "population": "N>=5 unique individuals, >=80% and >=5 people called per SNP; >=50000 shared qualifying SNPs for a supported pair.", "labels": "Explicit spelling synonyms reconciled; source subgroups preserved; AA abbreviation held unresolved.", "relatedness": "Relatedness flags are reported separately; distinct relatives are not duplicates and remain subject to the existing atlas policy." } }, "mongol_revision": { "retained_HGDP": 9, "excluded_nonHGDP": 83 }, "balkan_revision": [ { "population": "Bosnian", "N": 15, "qualifying_SNPs": 265927 }, { "population": "Croatian", "N": 34, "qualifying_SNPs": 91037 }, { "population": "Kosovar", "N": 9, "qualifying_SNPs": 265924 }, { "population": "Macedonian", "N": 14, "qualifying_SNPs": 265927 }, { "population": "Montenegrin", "N": 14, "qualifying_SNPs": 265927 }, { "population": "Serbian", "N": 18, "qualifying_SNPs": 265927 } ], "default_estimator": "Hudson with inferred single-allele adjustment for 214 zero-heterozygosity records", "ploidy_assumption": "Exactly zero heterozygous calls across the 302955-SNP panel implies one observed allele per called SNP for sensitivity-adjusted estimates. This inference is not an official AADR metadata correction.", "ploidy_validation": { "pairs": 50721, "finite_pairs": 50721, "blocks": 61, "rule": "For each pair, retain a SNP only when at least 80% of retained individuals AND at least 5 individuals have a genotype in EACH population. Count individuals, not allele copies. Apply the identical mask to adjusted and diploid estimates.", "minimum_pair_SNPs": 9370, "maximum_pair_SNPs": 302955, "negative_pairs": 6, "swedish_gbr": [ { "POP1": "GBR", "POP2": "Swedish", "N1": 82, "N2": 19, "OBS_CT": 267226, "HUDSON_FST": 0.0012929531443214098, "SE": 0.00013143324274943763, "DIPLOID_PLINK_FST": 0.0013518698415358882, "PLOIDY_SENSITIVITY_DELTA": -5.891669721447843e-05, "ASSUMED_HAPLOID_N1": 1, "ASSUMED_HAPLOID_N2": 0 } ], "swedish_gbr_sensitivity": [ { "POP1": "GBR", "POP2": "Swedish", "Fst_50": 0.0012939997873041332, "SNPs_50": 267289.0, "Fst_80": 0.0012929531443214098, "SNPs_80": 267226.0, "Fst_90": 0.0013004308607170343, "SNPs_90": 264842.0 } ], "diploid_note": "Hudson diploid calculation with coverage mask; compatibility column name DIPLOID_PLINK_FST retained, not a fresh PLINK run." }, "comparability": "Recomputed on a new sample and SNP panel; values replace rather than extend the old AADR matrix.", "uncertainty": "SE is a block jackknife over 5000 consecutive SNPs. It does not include population-label, ploidy-inference, ascertainment or geographic-provenance uncertainty. Marker overlap varies by pair. Negative estimates retained.", "references": [ { "citation": "Bhatia, G., Patterson, N., Sankararaman, S. & Price, A.L. (2013). Estimating and interpreting FST: The impact of rare variants. Genome Research, 23(9), 1514-1521.", "url": "https://doi.org/10.1101/gr.154831.113" }, { "citation": "Mallick, S. et al. (2024). The Allen Ancient DNA Resource (AADR) a curated compendium of ancient human genomes. Scientific Data, 11, 182.", "url": "https://doi.org/10.1038/s41597-024-03031-7" } ] }