What-if effects: A tale of entangled covariates
Zhaowei Yu, Yong Zhang · 2026
Read What-if effects: A tale of entangled covariates by Zhaowei Yu, Yong Zhang (2026) 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/pmc-13576651.txt.
Description
Zhaowei Yu and Yong Zhang spotlight a statistical trap in single-cell studies of human disease, and a new tool built to escape it. In real patient cohorts, the variables researchers care about travel together: disease with age, weight, medication, and sample quality. When everything is entangled, a molecular difference blamed on disease may really belong to a companion factor. The problem bites hardest in single-nucleus chromatin data, which are sparse and vast.
The commentary discusses DeepDive, a deep-learning framework from a recent study. It separates the known effects of each recorded factor from leftover variation in the data. That separation allows what-if experiments on paper: how would this cell's chromatin look without the disease, or without the drug. Yu and Zhang present the approach as a needed step toward honest causal reading of cohort-scale cell atlases.
Catalog record
| CREATOR | Zhaowei Yu, Yong Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 8 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |