Finemapping

Often at a region of genetic association with phenotype, there will be many variants that are statistically significant. Although, sometimes there are instances where one coding SNP is the only variant at the locus, these are quite rare overall.

Finemapping takes all the variants that are significant at a locus and assigns a probability that each is the causal variant. Any variants that are assigned a non-zero probability are then part of a "credible set", this credible set is displayed in PheWeb (refer to Additional Reading) and is also used in calculating colocalization.

Additional Reading:

• Finemapping methods used in FinnGen : from the FinnGen Release notes for R5, but same methods currently in use

• Finemapping results format in FinnGen

• How to access Finemapping data in FinnGen : by Mark Daly, Users' meeting tutorial

• Slides on the statistical background of Finemapping : From Matti Pirinen's GWAS course, start at Slide 21 for Finemapping.

• Detailed notes on finemapping from Matti Pirinen's GWAS course (page 23 starts Finemapping)

• Updates on FineMapping from Masa Kanai: in this Users' meeting tutorial, Masa describes some of the challenges in trans-ethnic finemapping and the filtering and processing updates that will become part of FinnGen