Negative binomial mixture model for identification of noise in antibody-antigen specificity predictions from single-cell data
Perry T Wasdin, Alexandra A Abu-Shmais, Michael W Irvin, Matthew J Vukovich, Ivelin S Georgiev, Magnus Rattray · 2024
Read Negative binomial mixture model for identification of noise in antibody-antigen specificity predictions from single-cell data by Perry T Wasdin, Alexandra A Abu-Shmais, Michael W Irvin, Matthew J Vukovich, Ivelin S Georgiev, Magnus Rattray (2024) 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-11631427.txt.
Description
Perry Wasdin and co-authors present a statistical cleanup method for a single-cell antibody screening technique in this 2024 paper. The technique reads which antigens a person's B cells bind, cell by cell, and the resulting counts mix true signal with technical noise. Cleaning that noise matters for finding real antibodies and for training prediction models.
Their method sorts antigen counts into signal and noise parts using a mixture model built for count data. It was tuned on a large multi-donor dataset from an earlier study. Tested on nine donors from separate runs, the cleaned data predicted lab-measured binding better than the standard scoring method. The gain held even though the runs differed in size and noise pattern. The authors expect the method to make such datasets more trustworthy as they grow.
Catalog record
| CREATOR | Perry T Wasdin, Alexandra A Abu-Shmais, Michael W Irvin, Matthew J Vukovich, Ivelin S Georgiev, Magnus Rattray |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |