Epigenetic modelling of former, current and never smokers

Clinical Epigenetics

Table 4 Performance of ternary classifiers of smoking status (current, former and never)

Data	Accuracy statistics	AHRR model (reference)	Candidate CpG LASSO model	Maas model	Agnostic LASSO model
Training data	Accuracy (95% CI)	0.606 (0.576–0.635)	0.538 (0.508–0.568)	0.619 (0.589–0.648)	0.695 (0.667–0.723)
	NIR (P: Acc > NIR)	0.364 (< 2.2 × 10⁻¹⁶)	0.364 (< 2.2 × 10⁻¹⁶)	0.364 (< 2.2 × 10⁻¹⁶)	0.364 (< 2.2 × 10⁻¹⁶)
	Kappa	0.405	0.306	0.427	0.541
	Never smokers
	Sensitivity	0.835	0.824	0.764	0.885
	Specificity	0.661	0.743	0.745	0.744
	PPV	0.562	0.625	0.610	0.643
	NPV	0.885	0.890	0.858	0.925
	Former smokers
	Sensitivity	0.299	0.380	0.455	0.518
	Specificity	0.872	0.763	0.797	0.892
	PPV	0.518	0.423	0.507	0.687
	NPV	0.731	0.729	0.762	0.802
	Current smokers
	Sensitivity	0.658	0.397	0.625	0.669
	Specificity	0.875	0.802	0.887	0.905
	PPV	0.730	0.512	0.743	0.787
	NPV	0.830	0.720	0.819	0.839
External validation data	Accuracy (95% CI)	0.612 (0.576–0.648)	0.594 (0.557–0.609)	0.603 (0.566–0.639)	0.637 (0.601–0.673)
	NIR (P: Acc > NIR)	0.367 (< 2.2 × 10⁻¹⁶)	0.367 (< 2.2 × 10⁻¹⁶)	0.367 (< 2.2 × 10⁻¹⁶)	0.367 (< 2.2 × 10⁻¹⁶)
	Kappa	0.405	0.390	0.406	0.462
	Never smokers
	Sensitivity	0.900	0.869	0.835	0.873
	Specificity	0.766	0.818	0.814	0.792
	PPV	0.686	0.731	0.719	0.705
	NPV	0.931	0.917	0.896	0.917
	Former smokers
	Sensitivity	0.171	0.368	0.297	0.270
	Specificity	0.914	0.811	0.824	0.916
	PPV	0.536	0.530	0.494	0.651
	NPV	0.656	0.689	0.669	0.684
	Current smokers
	Sensitivity	0.825	0.531	0.706	0.820
	Specificity	0.748	0.767	0.771	0.757
	PPV	0.548	0.458	0.533	0.556
	NPV	0.920	0.815	0.876	0.919

N.B. Ternary classifiers are the result of two binary classifiers being applied to DNAm data in sequence: ever versus never smoker classification, then current versus former classification of the ever smokers

ISSN: 1868-7083