From what I can see the analysis was not highly significant for the individual SNPs (p>0.1) but they combined it into a gene based analysis that brought the sig

From what I can see the analysis was not highly significant for the individual SNPs (p>0.1) but they combined it into a gene based analysis that brought the significance up. I'm not enough of a statistician to know how valid that procedure is but I would say even if you find out the individual SNPs it probably isn't too reliable to interpret them on their own.