# This README belongs to the summary statistics file "morning_person_BOLT.output_HRC.only_plus.metrics_maf0.001_hwep1em12_info0.3_logOR.txt.gz" and was generated for the secondary morning person GWA anaylsis described in "Genome-wide association analyses of chronotype in 697,828 individuals provides insights into circadian rhythms" (doi.org/10.1038/s41467-018-08259-7). The file contain 11 columns:
SNP - RSID of variant (or CHR:POS_ALLELE1_ALLELE0 if no RSID available)
CHR - chromosome number
BP - chromomsomal position (b37)
ALLELE1 - effect allele that the log odds ratio (LOGOR) corresponds to
ALLELE0 - non-effect allele
A1FREQ - frequency of ALLELE1
INFO - information/quality metric indicating whether a variant is well-imputed or not
LOGOR - (natural) log odds ratio for ALLELE1 on trait of interest
LOGOR_SE - standard error of log odds ratio
P_BOLT_LMM - non-infinitesimal mixed model association test p-value for this variant
HWE_P - p-value of Hardy-Weinberg equilibrium test performed in set of unrelated European individuals (~380,000 samples)
# For the morning person GWAS statistics, the LOGOR and LOGOR_SE have been calculated from BOLT-LMM's linear scale (as BOLT-LMM does not implement logistic regression and instead performs a linear regression). The conversion used is described in the BOLT-LMM manual at https://data.broadinstitute.org/alkesgroup/BOLT-LMM/#x1-5200010.2. The conversion for BETA to LOGOR is: LOGOR=BETA/(mu*(1-mu)), where mu = case fraction. The same conversion applies to the standard error: LOGOR_SE=SE/(mu*(1-mu)). For the morning person GWAS, the total number of samples was 403195 with 252287 cases (morning people) and 150908 controls (evening people), and with mu=(252287/403195)=.62572 (to 5 s.f.). If needed, the original BETA and SE columns can be calculated by rearranging the above equations.
# Only variants imputed to the Haplotype Reference Consortium and those with MAF>=0.1%, HWE-P>1E-12 and INFO>0.3 were included in our analyses and therefore in this summary data. Variants may have INFO<0.3 in this file as we filtered by INFO_UNR, calculated in the subset of unrelated Europeans (INFO_UNR not included in this file). The INFO provided in this file is calculated by BOLT-LMM using the 403195 European samples and includes related individuals.