Predicting ionizing radiation exposure using biochemically-inspired genomic machine learning

Jonathan Z L Zhao; Eliseos J Mucaki; Peter K Rogan

doi:10.12688/f1000research.14048.2

Predicting ionizing radiation exposure using biochemically-inspired genomic machine learning

F1000Res. 2018 Feb 27:7:233. doi: 10.12688/f1000research.14048.2. eCollection 2018.

Authors

Jonathan Z L Zhao^{1

2}, Eliseos J Mucaki¹, Peter K Rogan^{1

2

3

4

5}

Affiliations

¹ Department of Biochemistry, Schulich School of Medicine and Dentistry, Western University, London, ON, N6A 2C1, Canada.
² Department of Computer Science, Faculty of Science, Western University, London, ON, N6A 2C1, Canada.
³ Department of Epidemiology & Biostatistics, Schulich School of Medicine and Dentistry, Western University, London, ON, N6A 2C1, Canada.
⁴ CytoGnomix Inc., London, ON, N5X 3X5, Canada.
⁵ Department of Oncology, Schulich School of Medicine and Dentistry, Western University, London, ON, N6A 2C1, Canada.

Abstract

Background: Gene signatures derived from transcriptomic data using machine learning methods have shown promise for biodosimetry testing. These signatures may not be sufficiently robust for large scale testing, as their performance has not been adequately validated on external, independent datasets. The present study develops human and murine signatures with biochemically-inspired machine learning that are strictly validated using k-fold and traditional approaches. Methods: Gene Expression Omnibus (GEO) datasets of exposed human and murine lymphocytes were preprocessed via nearest neighbor imputation and expression of genes implicated in the literature to be responsive to radiation exposure (n=998) were then ranked by Minimum Redundancy Maximum Relevance (mRMR). Optimal signatures were derived by backward, complete, and forward sequential feature selection using Support Vector Machines (SVM), and validated using k-fold or traditional validation on independent datasets. Results: The best human signatures we derived exhibit k-fold validation accuracies of up to 98% ( DDB2, PRKDC, TPP2, PTPRE, and GADD45A) when validated over 209 samples and traditional validation accuracies of up to 92% ( DDB2, CD8A, TALDO1, PCNA, EIF4G2, LCN2, CDKN1A, PRKCH, ENO1, and PPM1D) when validated over 85 samples. Some human signatures are specific enough to differentiate between chemotherapy and radiotherapy. Certain multi-class murine signatures have sufficient granularity in dose estimation to inform eligibility for cytokine therapy (assuming these signatures could be translated to humans). We compiled a list of the most frequently appearing genes in the top 20 human and mouse signatures. More frequently appearing genes among an ensemble of signatures may indicate greater impact of these genes on the performance of individual signatures. Several genes in the signatures we derived are present in previously proposed signatures. Conclusions: Gene signatures for ionizing radiation exposure derived by machine learning have low error rates in externally validated, independent datasets, and exhibit high specificity and granularity for dose estimation.

Keywords: Biodosimetry; Gene Signatures; Ionizing Radiation Exposure; Machine Learning; Minimum Redundancy Maximum Relevance; Molecular Diagnostics; Support Vector Machine; Validation.

Grants and funding

Natural Sciences and Engineering Research Council of Canada (NSERC Discovery Grant RGPIN-2015-06290); the Canadian Foundation for Innovation; Canada Research Chairs, and CytoGnomix Inc.