CHIMA: a correlation-aware high-dimensional mediation analysis with its application to the living brain project study

Samuel Osarfo, Sangyoon Yi, Weijia Fu, Seungjun Ahn · 2025

Read CHIMA: a correlation-aware high-dimensional mediation analysis with its application to the living brain project study by Samuel Osarfo, Sangyoon Yi, Weijia Fu, Seungjun Ahn (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-16883.txt.

Description

Samuel Osarfo, Sangyoon Yi, Weijia Fu, and Seungjun Ahn improve a method for tracing how an exposure reaches an outcome through middlemen, in a 2025 paper. Such pathway analysis is called mediation analysis. Modern data holds thousands of candidate middlemen at once. Standard tools screen them, then test the survivors. When middlemen are strongly linked, screening slips and power falls.

Their method, CHIMA, adds two fixes. A projection step screens more accurately under strong links. A near-orthogonal cleanup step trims bias in the final estimates. Simulations show the method finds true middlemen even under heavy correlation. It beat rival methods across many settings. The team then applied it to brain gene data from the Living Brain Project. They found genes that carry the effect of Parkinson's disease onto brain cell makeup. The results reveal cell-type-specific routes of the disease.

Catalog record

CREATORSamuel Osarfo, Sangyoon Yi, Weijia Fu, Seungjun Ahn
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE49 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-16883

Browse the full library · About the Tower of Babel