University of Leicester
Browse
Comparative performances of machine learning methods for classifying Crohn Disease patients using genome-wide genotyping data.pdf (3.15 MB)

Comparative performances of machine learning methods for classifying Crohn Disease patients using genome-wide genotyping data.

Download (3.15 MB)
journal contribution
posted on 2019-10-02, 09:13 authored by Alberto Romagnoni, Simon Jégou, Kristel Van Steen, Gilles Wainrib, Jean-Pierre Hugot, International Inflammatory Bowel Disease Genetics Consortium (IIBDGC)
Crohn Disease (CD) is a complex genetic disorder for which more than 140 genes have been identified using genome wide association studies (GWAS). However, the genetic architecture of the trait remains largely unknown. The recent development of machine learning (ML) approaches incited us to apply them to classify healthy and diseased people according to their genomic information. The Immunochip dataset containing 18,227 CD patients and 34,050 healthy controls enrolled and genotyped by the international Inflammatory Bowel Disease genetic consortium (IIBDGC) has been re-analyzed using a set of ML methods: penalized logistic regression (LR), gradient boosted trees (GBT) and artificial neural networks (NN). The main score used to compare the methods was the Area Under the ROC Curve (AUC) statistics. The impact of quality control (QC), imputing and coding methods on LR results showed that QC methods and imputation of missing genotypes may artificially increase the scores. At the opposite, neither the patient/control ratio nor marker preselection or coding strategies significantly affected the results. LR methods, including Lasso, Ridge and ElasticNet provided similar results with a maximum AUC of 0.80. GBT methods like XGBoost, LightGBM and CatBoost, together with dense NN with one or more hidden layers, provided similar AUC values, suggesting limited epistatic effects in the genetic architecture of the trait. ML methods detected near all the genetic variants previously identified by GWAS among the best predictors plus additional predictors with lower effects. The robustness and complementarity of the different methods are also studied. Compared to LR, non-linear models such as GBT or NN may provide robust complementary approaches to identify and classify genetic markers.

Funding

This work was supported by Fondation pour la Recherche Médical (ref DEI20151234405) and Investissements d’Avenir programme ANR-11-IDEX-0005-02, Sorbonne Paris Cite, Laboratoire d’excellence INFLAMEX. The authors thank the students that participated to the wisdom of the crowd exercise.

History

Citation

Scientific Reports, 2019, 9, Article number: 10351

Author affiliation

/Organisation/COLLEGE OF LIFE SCIENCES/School of Medicine/Department of Health Sciences

Version

  • VoR (Version of Record)

Published in

Scientific Reports

Publisher

Nature Research (part of Springer Nature)

eissn

2045-2322

Acceptance date

2019-07-03

Copyright date

2019

Available date

2019-10-02

Publisher version

https://www.nature.com/articles/s41598-019-46649-z

Notes

Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-46649-z.

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC