Developing polygenic risk models for complex traits and diseases
entryage_bq
ph_X_bq (here, X= lung, breast, prostate, ovarian, colorectal, melanoma, pancreatic)
fstcan_exitstat
X_exitstat (X= lung, breast, prostate, ovarian, colorectal, melanoma, pancreatic)
age
Sex
X_cancer (X= lung, breast, prostate, ovarian, colorectal, melanoma, pancreatic)
X_stage (X= lung, breast, prostate, ovarian, colorectal, melanoma, pancreatic)
bq_age
race7
cig_stat
fh_cancer
bmi_curr
height_f
weight_f
arthrit_f
bronchit_f
colon_comorbidity
diabetes_f
divertic_f
emphys_f
gallblad_f
hearta_f
hyperten_f
liver_comorbidity
osteopor_f
polyps_f
stroke_f
lmenstr
benign_ovcyst
endometriosis
infpros_f
infprosa
prosprob_f
We have built PRS models using data from public GWAS summary statistics and would like to evaluate the performance using PLCO data (the listed traits) by calculating AUC. A separate application was submitted to access PLCO genotype data for calculating PRS.
Bin Zhu, BB/DCEG/NCI
Difei Wang, CGR/DCEG/NCI
Kevin Wang, CGR/DCEG/NCI
Xing Hua, CGR/DCEG/NCI