These datasets are from the study:

Choi SH, Jurgens SJ, et al. Sequencing in over 50,000 cases identifies coding and structural variants underlying atrial fibrillation risk. Nature Genetics, accepted for publication. 2023.

File descriptions:

Jurgens_2023_AF_WES_WGS_2022_summary_stats_genebased_LOFpDM_cMAC20_withFirth.tsv.gz
- Contains the summary stats for gene-based testing of rare LOF and predicted-deleterious missense variants. The 'raw' effect size corresponds to the effects estimated from score stats, while the Firth columns show Firth estimates for associations with some evidence of association. Results represent a meta-analysis of CCDG, TOPMed, UKBB and FOURIER. 

Jurgens_2023_AF_WES_WGS_2022_summary_stats_singlevariants_moderatehigh_MAC20.tsv.gz
- Contains summary stats for single variant testing of low-frequency and rare variants with HIGH or MODERATE VEP impact. We did not include the Firth estimates here, so all estimates are 'raw', i.e., estimated from the score statistics. Results represent a meta-analysis of CCDG, TOPMed, UKBB and FOURIER. 

Jurgens_2023_AF_WES_WGS_2022_summary_stats_genebased_SVc45_cMAC10_withFirth.tsv.gz
- Contains summary stats for gene-based testing of rare structural variants with predicted class 4 or 5 impact. Firth estimates are added for highly significant genes. Note: results are only from TOPMed.