Bi-directional causal effects were tested between lifetime cannabis use and schizophrenia. We used genetic variants from our cannabis GWAS as well as those from the largest schizophrenia GWAS66 to serve as instruments (gene-exposure association). For lifetime cannabis use we used 2 genetic instruments; 1) an instrument including all independent genetic variants that were genome-wide significantly associated with lifetime cannabis use (p<5e-08; 5 SNPs), and 2) an instrument including independent variants with a more lenient significance threshold (p<1e-05; 69 SNPs). For schizophrenia we used one genetic instrument, including independent genetic variants that were genome-wide significantly associated with schizophrenia (instrument p<5e-08; 109 SNPs). Information on the included SNPs in the genetic instruments is provided in Supplementary Table S7.
