Inferring correlation networks from genomic survey data.
Friedman J, Alm EJ. · 2012
Read Inferring correlation networks from genomic survey data. by Friedman J, Alm EJ. (2012) 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/europmc-pmc3447976.txt.
Description
Friedman and Alm exposed a statistical trap in 2012 lurking in microbial survey data, then built a tool to escape it. Surveys report each species as a share of the whole, not a headcount, and correlation on shares is notoriously unreliable, a warning going back to Pearson. Using simulated data and Human Microbiome Project samples, they showed how badly the standard approach misbehaves.
Many apparent links between species were artifacts, and some true links appeared with the wrong sign. Diverse communities softened the distortion; sparse ones suffered most. Their method, SparCC, estimates true correlations from share data anyway. Judged against its network, the standard approach invented roughly three false links per true one in gut data and missed most real ones.
Catalog record
| CREATOR | Friedman J, Alm EJ. |
|---|---|
| YEAR | 2012 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |