Quantitative prediction of selectivity between the A1 and A2A adenosine receptors

Burggraaff, Lindsey; van Vlijmen, Herman W. T.; IJzerman, Adriaan P.; van Westen, Gerard J. P.

doi:10.1186/s13321-020-00438-3

Journal of Cheminformatics

Table 4 Performances of selectivity modeling using the two-step A₁AR-A_2AAR difference model or the selectivity-window model

From: Quantitative prediction of selectivity between the A₁ and A_2A adenosine receptors

Model	Class	MCC	Sensitivity	Specificity	PPV	NPV	ROC
Trained on all double points, tested on all double points (non-binders and semi-selective compounds always true/false negative)
A₁AR-A_2AAR difference	A₁AR	0.15 ± 0.13	0.17 ± 0.14	0.99 ± 0.00	0.18 ± 0.12	0.98 ± 0.01	0.76 ± 0.09
	A_2AAR	0.11 ± 0.07	0.19 ± 0.09	0.94 ± 0.02	0.15 ± 0.07	0.94 ± 0.01	0.67 ± 0.14
	Dual	0.26 ± 0.05	0.76 ± 0.02	0.50 ± 0.07	0.68 ± 0.02	0.59 ± 0.03	0.66 ± 0.02
Selectivity-window	A₁AR	0.07** ± 0.09	0.030.03	0.99 ± 0.00	0.10** ± 0.10	0.97 ± 0.00	0.87 ± 0.03
	A_2AAR	0.22 ± 0.12	0.15 ± 0.09	0.99 ± 0.00	0.42 ± 0.18	0.94 ± 0.01	0.74 ± 0.07
	Dual	0.36 ± 0.06	0.85 ± 0.03	0.48 ± 0.03	0.69 ± 0.01	0.70 ± 0.06	0.70 ± 0.02
Trained on all double points, tested on only A₁AR, A_2AAR, and dual class
A₁AR-A_2AAR difference	A₁AR	0.19 ± 0.16	0.17 ± 0.14	0.99 ± 0.00	0.28 ± 0.18	0.96 ± 0.01	0.75 ± 0.09
	A_2AAR	0.28* ± 0.12	0.19 ± 0.09	0.98 ± 0.01	0.48* ± 0.15	0.91 ± 0.01	0.72 ± 0.15
	Dual	0.25 ± 0.07	0.76 ± 0.02	0.57 ± 0.11	0.91 ± 0.02	0.28 ± 0.04	0.70 ± 0.05
Selectivity-window	A₁AR	0.17** ± 0.23	0.03 ± 0.03	0.99 ± 0.01	0.50** ± 0.50	0.92 ± 0.01	0.88 ± 0.04
	A_2AAR	0.32* ± 0.15	0.15 ± 0.09	1.00 ± 0.00	0.75* ± 0.25	0.82 ± 0.03	0.80 ± 0.07
	Dual	0.33 ± 0.11	0.84 ± 0.04	0.46 ± 0.06	0.80 ± 0.02	0.55 ± 0.10	0.66 ± 0.04
Trained on only A₁AR, A_2AAR, and dual class, tested on only A₁AR, A_2AAR, and dual class
Selectivity-window	A₁AR	− 0.05*** ± 0.00	0.00 ± 0.00	0.99 ± 0.01	0.00*** ± 0.00	0.92 ± 0.01	0.81 ± 0.05
	A_2AAR	0.23 ± 0.18	0.25 ± 0.12	0.94 ± 0.03	0.43 ± 0.22	0.83 ± 0.03	0.66 ± 0.11
	Dual	0.04 ± 0.12	0.69 ± 0.05	0.35 ± 0.09	0.73 ± 0.03	0.32 ± 0.09	0.53 ± 0.08

The class indicates the performance for that particular selectivity class: A₁AR-selective, A_2AAR-selective, and dual (non-selective). The query compounds were categorized based on post-classification of the selectivity predictions: A₁AR-selective when pActivity difference ≥ 2, A_2AAR-selective when pActivity difference ≤ − 2, and dual binder when pActivity difference ≥ − 1 and ≤ 1
MCC Matthews Correlation Coefficient, PPV positive predictive value, NPV negative predictive value, ROC receiver operating characteristic
*1/5 folds failed, **3/5 folds failed, ***4/5 folds failed

Back to article page

ISSN: 1758-2946

Contact us

Submission enquiries: journalsubmissions@springernature.com