- Q: We’ve successfully run the S-PrediXcan and S-MultiXcan methods on our GWAS data using the MASHR sQTL models and have the outputs. My question is, how do I interpret and/or follow up on the results? I see a handful of intron IDs that look significant and I find the relevant single tissue results for those intron IDs.
Find the mapping from intron ids to gene ids in this link https://uchicago.box.com/s/xy71r0su6refrfggivrwc7kpvsxysmmn
Q: How should I interpret the z-score? Does a negative z-score for an intron imply that a decrease in the excision of that intron leads to an increase in the GWAS trait? Yes
Q: How can I recover the SNPs that are predicted to lead to a decrease in the excision of that intron?
Use this post to query the database file where the prediction weights are included https://predictdb.org/post/2021/07/21/how-do-i-query-the-prediction-model-db/