Prediction of Smoking Behavior From Single Nucleotide Polymorphisms With Machine Learning Approaches

Abstract

Smoking is a complex behavior with a heritability as high as 50%. Given such a large genetic contribution, it provides an opportunity to prevent those individuals who are susceptible to smoking dependence from ever starting to smoke by predicting their inherited predisposition with their genomic profiles. Although previous studies have identified many susceptibility variants for smoking, they have limited power to predict smoking behavior. We applied the support vector machine (SVM) and random forest (RF) methods to build prediction models for smoking behavior. We first used 1,431 smokers and 1,503 non-smokers of African origin for model building with a 10-fold cross-validation and then tested the prediction models on an independent dataset consisting of 213 smokers and 224 non-smokers. The SVM model with 500 top single nucleotide polymorphisms (SNPs) selected using logistic regression (p<0.01) as the feature selection method achieved an area under the curve (AUC) of 0.691, 0.721, and 0.720 for the training, test, and independent test samples, respectively. The RF model with 500 top SNPs selected using logistic regression (p<0.01) achieved AUCs of 0.671, 0.665, and 0.667 for the training, test, and independent test samples, respectively. Finally, we used the combined logistic (p<0.01) and LASSO (l=10 −3) regression to select features and the SVM algorithm for model building. The SVM model with 500 top SNPs achieved AUCs of 0.756, 0.776, and 0.897 for the training, test, and independent test samples, respectively. We conclude that machine learning methods are promising means to build predictive models for smoking.

A conserved role of α2δ subunit of calcium channel in nicotine motivated behavior

“…Although GWAS studies associated with tobacco smoking have revealed numerous genetic factors, the estimated heritability has been limited to explaining underlying mechanisms. Thus, various attempts have been suggested to accelerate functional validation and comprehensive analysis 64 . In a comparative proteomics study, 83% of the worm proteome exhibits homology with human genes and recent meta-analysis with orthology-prediction methods showed that approximately 52.6% of the human protein-coding genome has noticeable orthologues in worms, illustrating that the nematode provides a suitable model organism for functional validation of human genes.…”

Prediction of prokaryotic transposases from protein features with machine learning approaches

“…Therefore, we further selected a set of features with satisfying predictive performances for Tnp prediction. ML approaches have been widely used in the selection of markers and construction of prediction models, and have been shown to improve the predictive performance of models in various human diseases [38, 39] and protein identification tasks [17, 40]. In this study, we applied a strategy of combining both MI and LASSO methods to reduce the dimension of protein features and finally selected 75 features as signatures for predicting Tnps.…”

Identification of a Novel Functional Non-synonymous Single Nucleotide Polymorphism in Frizzled Class Receptor 6 Gene for Involvement in Depressive Symptoms

“…Questionnaires assessing various other characteristics of interest were also administered, and individuals exhibiting other substance abuse (except smoking) or other psychiatric disorders (except depression) were excluded. As previously reported ( Yang et al, 2015 ; Yang and Li, 2016 ; Jiang et al, 2019 ; Xu et al, 2020 ), genomic DNA was isolated from the peripheral blood of each patient, and genotyping was performed using the Illumina Infinium HumanExome BeadChip v 1.0 (Illumina Inc., San Diego, CA, United States). This exome chip includes more than 240,000 functional exonic variants that have been detected in several sequencing data sets, and it was designed to concentrate on rare variants.…”