Abstract
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.
Highlights
Crohn Disease (CD) is an inflammatory bowel disease (IBD) characterized by a chronic or relapsing inflammation of the gut with a prevalence of at least 0.1% in most developed countries[1]
Genetic variants consisted in biallelic SNPs and a few small insertion deletions polymorphisms25. 156499 variants survived after a first quality control (QC) performed according to the international consortium[3]
In this paper we considered three classes of models for case/control classification: logistic regression (LR), dense neural networks (NN) and gradient boosting on decision trees (GBT)
Summary
Crohn Disease (CD) is an inflammatory bowel disease (IBD) characterized by a chronic or relapsing inflammation of the gut with a prevalence of at least 0.1% in most developed countries[1]. Most associated variants are common with minor allele frequency (MAF) > 0:01) and risk alleles are either the minor or the major alleles Their effect sizes are usually small (OR < 1:5) and often smaller (from 1:1 and to 1:2). Until now the genomic information has mainly been exploited on the basis of single-locus statistical analyses This approach is under-powered to detect variants carrying low marginal effects alone but strong effects in association with other ones. Standard methods applied to higher orders of interactions were quickly limited by the multi-testing issue, the size of the datasets and computing power For these reasons, more sophisticated machine learning (ML) methods have been proposed in order to capture the whole information of GWAS datasets using a direct pan-genomic approach. To explore the performances of different methods, the receiver operator characteristic (ROC) curve and its maximum Area Under Curve (AUC) are often used to compare the sensitivity and specificity of genetic tests in correctly classifying affected and unaffected individuals[2]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.