Relief is a statistical method to perform feature selection. It could be used,

for instance, to find genomic loci that correlate with a trait or genes whoseexpression correlate with a condition. Relief can also be made sensitive tointeraction effects (known in genetics as epistasis).

In this episode, Trang Lê joins me

to talk about Relief and her version of Relief called STIR (STatisticalInference Relief). While traditional Relief algorithms could only rankfeatures and needed a user-supplied threshold to decide which features toselect, Trang’s reformulation of Relief allowed her to compute p-valuesand make the selection process less arbitrary.

Links:

• Paper: STatistical Inference Relief (STIR) feature selection
• STIR on GitHub
• Relief on Wikipedia
• The original Relief paper by Kira and Rendell (1992)
• Epistasis: what it means, what it doesn’t mean, and statistical methods to detect it in humans