Ge-SAND: an explainable deep learning-driven framework for disease risk prediction by uncovering complex genetic interactions in parallel

Ye, Lihang; Zhang, Liubin; Tang, Bin; Liang, Junhao; Tan, Ruijie; Jiang, Hui; Peng, Wenjie; Lin, Nan; Li, Kun; Xue, Chao; Li, Miaoxin

doi:10.1186/s12864-025-11588-9

Table 4 The performance of different methods in six models with 5,000 sample size

From: Ge-SAND: an explainable deep learning-driven framework for disease risk prediction by uncovering complex genetic interactions in parallel

Sample Size	Model	Method	AUC-ROC	AUC-PR	KS	MCC	ACC	Recall	Precision	F1 Score
5,000	LN	Ge-SAND	0.664	0.649	0.272	0.276	0.636	0.720	0.616	0.664
		LASSO	0.645	0.635	0.256	0.254	0.626	0.568	0.643	0.603
		Ridge	0.600	0.590	0.192	0.190	0.594	0.672	0.581	0.623
		Random Forest	0.586	0.566	0.136	0.136	0.568	0.604	0.563	0.583
		SVM	0.594	0.581	0.184	0.181	0.590	0.544	0.599	0.570
		XGBoost	0.652	0.643	0.268	0.266	0.632	0.572	0.650	0.609
		Naive Bayes	0.576	0.581	0.116	0.138	0.558	0.288	0.626	0.395
		MLP	0.585	0.598	0.192	0.200	0.596	0.460	0.632	0.532
		CNN	0.516	0.523	0.052	0.052	0.526	0.588	0.523	0.554
		LSTM	0.600	0.577	0.184	0.184	0.592	0.612	0.588	0.600
	QD	Ge-SAND	0.689	0.670	0.304	0.300	0.650	0.660	0.647	0.654
		LASSO	0.667	0.665	0.256	0.273	0.626	0.432	0.706	0.536
		Ridge	0.631	0.626	0.192	0.208	0.594	0.380	0.664	0.484
		Random Forest	0.554	0.563	0.120	0.124	0.560	0.436	0.580	0.498
		SVM	0.626	0.626	0.180	0.184	0.588	0.736	0.568	0.641
		XGBoost	0.641	0.643	0.216	0.236	0.608	0.408	0.680	0.510
		Naive Bayes	0.579	0.580	0.140	0.140	0.570	0.564	0.571	0.567
		MLP	0.608	0.595	0.184	0.184	0.592	0.588	0.593	0.590
		CNN	0.522	0.526	0.080	0.085	0.540	0.368	0.561	0.444
		LSTM	0.500	0.510	0.056	0.056	0.528	0.580	0.525	0.551
	LN + QD	Ge-SAND	0.717	0.701	0.340	0.360	0.668	0.848	0.624	0.719
		LASSO	0.681	0.693	0.288	0.292	0.644	0.724	0.624	0.670
		Ridge	0.666	0.685	0.260	0.265	0.630	0.532	0.662	0.590
		Random Forest	0.636	0.606	0.212	0.226	0.606	0.432	0.663	0.523
		SVM	0.655	0.674	0.260	0.261	0.628	0.528	0.660	0.587
		XGBoost	0.690	0.666	0.304	0.310	0.650	0.524	0.701	0.600
		Naive Bayes	0.639	0.662	0.252	0.287	0.624	0.372	0.750	0.497
		MLP	0.643	0.648	0.248	0.253	0.624	0.720	0.604	0.657
		CNN	0.509	0.504	0.060	0.063	0.530	0.672	0.523	0.588
		LSTM	0.497	0.490	0.040	0.039	0.516	0.808	0.510	0.625
	CB	Ge-SAND	0.697	0.647	0.324	0.347	0.662	0.840	0.620	0.713
		LASSO	0.655	0.633	0.280	0.294	0.638	0.808	0.603	0.691
		Ridge	0.624	0.586	0.220	0.221	0.610	0.664	0.599	0.630
		Random Forest	0.599	0.590	0.168	0.176	0.584	0.432	0.621	0.509
		SVM	0.624	0.594	0.196	0.220	0.598	0.824	0.568	0.672
		XGBoost	0.669	0.627	0.288	0.285	0.642	0.680	0.632	0.655
		Naive Bayes	0.624	0.588	0.224	0.221	0.610	0.664	0.599	0.630
		MLP	0.607	0.570	0.212	0.216	0.606	0.700	0.589	0.640
		CNN	0.509	0.513	0.060	0.062	0.530	0.416	0.539	0.470
		LSTM	0.522	0.530	0.060	0.062	0.530	0.664	0.524	0.586
	LN + CB	Ge-SAND	0.711	0.686	0.328	0.324	0.662	0.660	0.663	0.661
		LASSO	0.688	0.683	0.288	0.284	0.642	0.648	0.640	0.644
		Ridge	0.640	0.634	0.216	0.218	0.608	0.544	0.624	0.581
		Random Forest	0.593	0.591	0.140	0.141	0.570	0.508	0.580	0.542
		SVM	0.635	0.634	0.232	0.235	0.616	0.536	0.638	0.583
		XGBoost	0.689	0.680	0.284	0.294	0.642	0.512	0.692	0.589
		Naive Bayes	0.609	0.614	0.172	0.177	0.586	0.700	0.570	0.628
		MLP	0.640	0.639	0.220	0.224	0.610	0.516	0.635	0.570
		CNN	0.530	0.559	0.108	0.136	0.554	0.252	0.636	0.361
		LSTM	0.533	0.550	0.072	0.075	0.536	0.668	0.528	0.590
	QD + CB	Ge-SAND	0.746	0.748	0.364	0.360	0.680	0.672	0.683	0.677
		LASSO	0.693	0.692	0.308	0.308	0.652	0.576	0.679	0.623
		Ridge	0.656	0.664	0.236	0.242	0.616	0.760	0.590	0.664
		Random Forest	0.599	0.604	0.152	0.171	0.576	0.804	0.552	0.655
		SVM	0.644	0.647	0.228	0.229	0.614	0.580	0.622	0.600
		XGBoost	0.693	0.665	0.284	0.281	0.640	0.596	0.654	0.623
		Naive Bayes	0.588	0.590	0.136	0.158	0.568	0.312	0.639	0.419
		MLP	0.637	0.632	0.224	0.239	0.612	0.784	0.583	0.669
		CNN	0.467	0.482	0.084	0.011	0.504	0.168	0.512	0.253
		LSTM	0.493	0.496	0.076	0.128	0.538	0.940	0.521	0.670

Back to article page

ISSN: 1471-2164

Contact us

Submission enquiries: bmcgenomics@biomedcentral.com
General enquiries: ORSupport@springernature.com

BMC Genomics

Contact us