Welcome to the github page for the Leveraging supervised learning for functionally informed fine-mapping of cis-eQTLs identifies an additional 20,913 putative causal eQTLs. Nat Commun 12, 3394 (2021). paper, which introduces the Expression Modifier Score (EMS) and its applications to functionally-informed fine-mapping and co-localization.
The tutorial.ipynb
explains how to annotate EMS in your own dataset using hail in a cloud.
Since most of the analysis was performed in hail, we recommend users who are not familiar with Hail to visit the Hail tutorial page.
(Caution: the EMS dataset in google cloud are available at requiester-pays bucket. For more information, please visit the file description page)
For those who are not familiar with hail, he tutorial_local.ipynb
explains how to annotate EMS in local (EMS for a subset of variant is also available on our dropbox).
The scripts used in the EMS paper are stored in the code
directory.