Iterative random forests to discover predictive and stable high-order interactions

Journal Article · 18 January 2018 · Proceedings of the National Academy of Sciences of the United States of America

DOI: https://doi.org/10.1073/pnas.1711236115 · OSTI ID:1417528

Basu, Sumanta ^[1]; Kumbier, Karl ^[2]; Brown, James B. ^[3]; Yu, Bin ^[4]

---

1. Department of Biological Statistics and Computational Biology, Cornell University, Ithaca, NY 14853,, Department of Statistical Science, Cornell University, Ithaca, NY 14853,, Data Driven Decisions Department, Preminon LLC, Antioch, CA 94531,
2. Statistics Department, University of California, Berkeley, CA 94720,
3. Data Driven Decisions Department, Preminon LLC, Antioch, CA 94531,, Statistics Department, University of California, Berkeley, CA 94720,, Centre for Computational Biology, School of Biosciences, University of Birmingham, Edgbaston B15 2TT, United Kingdom,, Molecular Ecosystems Biology Department, Biosciences Area, Lawrence Berkeley National Laboratory, Berkeley, CA 94720,
4. Data Driven Decisions Department, Preminon LLC, Antioch, CA 94531,, Statistics Department, University of California, Berkeley, CA 94720,, Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA 94720

Significance We developed a predictive, stable, and interpretable tool: the iterative random forest algorithm (iRF). iRF discovers high-order interactions among biomolecules with the same order of computational cost as random forests. We demonstrate the efficacy of iRF by finding known and promising interactions among biomolecules, of up to fifth and sixth order, in two data examples in transcriptional regulation and alternative splicing.

• Sponsoring Organization: USDOE
• Grant/Contract Number: SC0017069
• OSTI ID: 1417528
• Journal Information: Proceedings of the National Academy of Sciences of the United States of America, Journal Name: Proceedings of the National Academy of Sciences of the United States of America Journal Issue: 8 Vol. 115; ISSN 0027-8424
• Publisher: Proceedings of the National Academy of SciencesCopyright Statement
• Country of Publication: United States
• Language: English