Semi-Supervised Methods to Predict Patient Survival from Gene Expression Data

Eric Bair, Robert Tibshirani · 2004

Read Semi-Supervised Methods to Predict Patient Survival from Gene Expression Data by Eric Bair, Robert Tibshirani (2004) 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/plos-pbio-0020108.txt.

Description

Eric Bair and Robert Tibshirani presented a statistics method in PLOS Biology in 2004. Doctors want to predict how long cancer patients will live using gene activity data. The catch is that such data sets hold thousands of genes but few patients, so plain methods often fail.

Their semi-supervised approach first finds gene patterns tied to outcome without using survival times, then links those patterns to survival. Splitting the task this way cuts the noise that swamps direct methods. Better survival forecasts help doctors sort patients by risk and choose how hard to treat. The paper is a good example of math built for the exact shape of a real data problem.

Catalog record

CREATOREric Bair, Robert Tibshirani
YEAR2004
COLLECTIONPAPERS
FORMATTXT
FILE SIZE69 KB
LICENSECreative Commons Attribution License
RIGHTSCC BY
SOURCEPLOS ONE

Source: https://journals.plos.org/plosone/article?id=10.1371/journal.pbio.0020108

Browse the full library · About the Tower of Babel