Predicted Trait | |
Reported Trait | Prostate carcinoma |
Mapped Trait(s) | prostate carcinoma (EFO_0001663) |
Score Construction | |
PGS Name | LDpred2 |
Development Method | |
Name | LDpred2 |
Parameters | LDpred2 estimates the posterior mean effect size of each SNP, while accounting for the effect estimates of other correlated SNPs. This method employs a point-normal prior, selecting only a subset of SNPs with non-zero estimated effects for inclusion in the PGS. LDpred2 involves three key hyperparameters: the fraction of causal SNPs, SNP-based heritability, and sparsity (True or False). Settings were selected based on their ability to produce the highest receiver operating characteristic curve (AUC) in the UKB tuning data, with final settings of 0.018 for the fraction of causal SNPs, 0.0866 for SNP-based heritability (70% of the LDSC estimate), and no sparsity, yielding an AUC of 0.7 in the UKB tuning data. |
Variants | |
Original Genome Build | GRCh37 |
Number of Variants | 964,607 |
Effect Weight Type | beta |
PGS Source | |
PGS Catalog Publication (PGP) ID | PGP000741 |
Citation (link to publication) | Tanha HM et al. HGG Adv (2025) |
Study Identifiers | Sample Numbers | Sample Ancestry | Cohort(s) |
---|---|---|---|
GWAS Catalog: GCST011049 Europe PMC: 33398198 |
177,526 individuals | European | NR |
GWAS Catalog: GCST011049 Europe PMC: 33398198 |
21,354 individuals | African American or Afro-Caribbean,African unspecified | NR |
GWAS Catalog: GCST011049 Europe PMC: 33398198 |
27,420 individuals | East Asian | NR |
GWAS Catalog: GCST011049 Europe PMC: 33398198 |
7,953 individuals | Hispanic or Latin American | NR |
Study Identifiers | Sample Numbers | Sample Ancestry | Cohort(s) | Phenotype Definitions & Methods | Age of Study Participants | Participant Follow-up Time | Additional Ancestry Description | Additional Sample/Cohort Information |
---|---|---|---|---|---|---|---|---|
— | [ ,
100.0 % Male samples |
European | UKB | Linked cancer registry records were used to identify participants diagnosed with malignant neoplasm of the prostate (ICD-10 code: C61; ICD-9 code: 185). | — | PC1-PC4 values ±3s.d. from the 1000G European cluster | — |
PGS Performance Metric ID (PPM) |
PGS Sample Set ID (PSS) |
Performance Source | Trait |
PGS Effect Sizes (per SD change) |
Classification Metrics | Other Metrics | Covariates Included in the Model |
PGS Performance: Other Relevant Information |
---|---|---|---|---|---|---|---|---|
PPM022676 | PSS012055| African Ancestry| 3,193 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.592 [0.533, 0.651] | — | - | — |
PPM022671 | PSS012056| European Ancestry| 184,010 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.698 [0.689, 0.707] | — | - | — |
PPM022681 | PSS012057| South Asian Ancestry| 5,097 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.742 [0.664, 0.82] | — | - | — |
PPM022686 | PSS012054| European Ancestry| 6,791 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.702 [0.67, 0.733] | — | - | — |
PPM022691 | PSS012053| European Ancestry| 1,809 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.717 [0.687, 0.747] | — | - | — |
PPM022696 | PSS012056| European Ancestry| 184,010 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.789 [0.782, 0.796] | — | Age-specific absolute risk adjusted by PGS relative risk | — |
PPM022701 | PSS012057| South Asian Ancestry| 5,097 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.842 [0.79, 0.894] | — | Age-specific absolute risk adjusted by PGS relative risk | — |
PPM022706 | PSS012055| African Ancestry| 3,193 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.798 [0.759, 0.838] | — | Age-specific absolute risk adjusted by PGS relative risk | — |
PPM022716 | PSS012053| European Ancestry| 1,809 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.725 [0.696, 0.754] | — | Age-specific absolute risk adjusted by PGS relative risk | — |
PPM022711 | PSS012054| European Ancestry| 6,791 individuals |
PGP000741 | Tanha HM et al. HGG Adv (2025) |
Reported Trait: 5-year incident prostate cancer | — | AUROC: 0.741 [0.714, 0.768] | — | Age-specific absolute risk adjusted by PGS relative risk | — |
PGS Sample Set ID (PSS) |
Phenotype Definitions and Methods | Participant Follow-up Time | Sample Numbers | Age of Study Participants | Sample Ancestry | Additional Ancestry Description | Cohort(s) | Additional Sample/Cohort Information |
---|---|---|---|---|---|---|---|---|
PSS012053 | Linked cancer registry records were used to identify participants diagnosed with incident malignant neoplasm of the prostate (ICD-10 code: C61; ICD-9 code: 185) within five years of prediction baseline. | — | [ ,
100.0 % Male samples |
European | PC1-PC4 values ±3s.d. from the 1000G European cluster | MCCS | MCCS had a case/sub-cohort design, leading to a higher proportion of cases | |
PSS012054 | Linked cancer registry records were used to identify participants diagnosed with incident malignant neoplasm of the prostate (ICD-10 code: C61; ICD-9 code: 185) within five years of prediction baseline. | — | [ ,
100.0 % Male samples |
European | PC1-PC4 values ±3s.d. from the 1000G European cluster | QSKIN | — | |
PSS012055 | Linked cancer registry records were used to identify participants diagnosed with incident malignant neoplasm of the prostate (ICD-10 code: C61; ICD-9 code: 185) within five years of prediction baseline. | — | [ ,
100.0 % Male samples |
African American or Afro-Caribbean | AFR individuals were identified by PC1-PC4 from 1000G and k-means clustering. | UKB | — | |
PSS012056 | Linked cancer registry records were used to identify participants diagnosed with incident malignant neoplasm of the prostate (ICD-10 code: C61; ICD-9 code: 185) within five years of prediction baseline. | — | [ ,
100.0 % Male samples |
European | EUR individuals were identified by PC1-PC4 from 1000G and k-means clustering. | UKB | — | |
PSS012057 | Linked cancer registry records were used to identify participants diagnosed with incident malignant neoplasm of the prostate (ICD-10 code: C61; ICD-9 code: 185) within five years of prediction baseline. | — | [ ,
100.0 % Male samples |
South Asian | SAS individuals were identified by PC1-PC4 from 1000G and k-means clustering. | UKB | — |