Table 1. Characteristics of the studies included for the gene–iron and gene–PD associations.

Data Source	n studies	Type of Study	Maximum Sample Size
Gene–iron association
GIS Consortiuma	10	GWA	21,567
Gene–PD association
PDGene database [13]–[21]	9	Candidate gene studies (HFE rs1800562 and HFE rs1799945)	2,384 cases; 6,908 controls
PD GWAS Consortium [22]	5	GWA	4,238 cases; 4,239 controls
23andMe [23] b	1	GWA	4,127 cases; 62,037 controls
IPDGC [24],[25] c	4	GWA	4,258 cases; 10,152 controls
IPDGC [24],[25]	5	Immunochip genotyping	5,802 cases; 5,556 controls

Details on individual datasets are reported in Text S1 and in Tables S1 and S2.

^aUnpublished data. The original sample size was 22,444, but genotype and phenotype data were available only for 21,567.

^b23andMe: slightly expanded version of the cohort used in [23].

^cIPDGC: USA-NIA and USA-dbGAP studies were not included in our analysis due to overlap with PD GWAS Consortium; the Icelandic dataset was not available for analysis.