Limits...
Artificial neural network cascade identifies multi-P450 inhibitors in natural compounds.

Li Z, Li Y, Sun L, Tang Y, Liu L, Zhu W - PeerJ (2015)

Bottom Line: Thus, multi-P450 inhibition leads to greater drug-drug interaction risk than specific P450 inhibition.The results indicate significant positive correlation between the PIS values and the number of inhibited P450 isoforms (Spearman's ρ = 0.684, p < 0.0001).Furthermore, chemical similarity calculations suggested that the prevailing parent structures of natural multi-P450 inhibitors were alkaloids.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Pharmacy Administration, Harbin Medical University , Harbin , China.

ABSTRACT
Substantial evidence has shown that most exogenous substances are metabolized by multiple cytochrome P450 (P450) enzymes instead of by merely one P450 isoform. Thus, multi-P450 inhibition leads to greater drug-drug interaction risk than specific P450 inhibition. Herein, we innovatively established an artificial neural network cascade (NNC) model composed of 23 cascaded networks in a ladder-like framework to identify potential multi-P450 inhibitors among natural compounds by integrating 12 molecular descriptors into a P450 inhibition score (PIS). Experimental data reporting in vitro inhibition of five P450 isoforms (CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4) were obtained for 8,148 compounds from the Cytochrome P450 Inhibitors Database (CPID). The results indicate significant positive correlation between the PIS values and the number of inhibited P450 isoforms (Spearman's ρ = 0.684, p < 0.0001). Thus, a higher PIS indicates a greater possibility for a chemical to inhibit the enzyme activity of at least three P450 isoforms. Ten-fold cross-validation of the NNC model suggested an accuracy of 78.7% for identifying whether a compound is a multi-P450 inhibitor or not. Using our NNC model, 22.2% of the approximately 160,000 natural compounds in TCM Database@Taiwan were identified as potential multi-P450 inhibitors. Furthermore, chemical similarity calculations suggested that the prevailing parent structures of natural multi-P450 inhibitors were alkaloids. Our findings show that dissection of chemical structure contributes to confident identification of natural multi-P450 inhibitors and provides a feasible method for virtually evaluating multi-P450 inhibition risk for a known structure.

No MeSH data available.


Related in: MedlinePlus

Ten-Fold cross-validation of NNC model II and ANN model II.(A) The AUROCs are 0.864 and 0.842 for discrimination between P450 inhibitors(n = 1–5) and P450 non-inhibitors (n = 0) usingNNC model II and ANN model II, respectively. (B) The AUROCs are 0.845 and 0.822 foridentification of non-multi-P450 inhibitors (n = 0–2) andmulti-P450 inhibitors (n = 3–5) using the two models,respectively.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4696407&req=5

fig-4: Ten-Fold cross-validation of NNC model II and ANN model II.(A) The AUROCs are 0.864 and 0.842 for discrimination between P450 inhibitors(n = 1–5) and P450 non-inhibitors (n = 0) usingNNC model II and ANN model II, respectively. (B) The AUROCs are 0.845 and 0.822 foridentification of non-multi-P450 inhibitors (n = 0–2) andmulti-P450 inhibitors (n = 3–5) using the two models,respectively.

Mentions: The holdout cross-validation method was used for internal validation of each ANN submodelin the two NNC models and the two ANN models. Similar values ofRTe and RTr guaranteedsatisfactory generalizability of the constructed models (Figs. 2 and 3). A set of 2,716 compoundswith complete in vitro P450 inhibition data was applied to test NNC modelI and ANN model I for method validation. The PIS values exported from the two models weresignificantly positively corrected with the normalized number of inhibited P450 isoforms(Spearman’s rho = 0.613 and 0.587 for NNC model I and ANN model I, respectively,p < 0.0001). For NNC model II and ANN model II, the 10-foldcross-hold method was used for internal validation. Significant correlations between thePIS scores and the normalized number of inhibited P450 isoforms were observed for bothmodels (Spearman’s rho = 0.686 and 0.645 for NNC model II and ANN model II, respectively,p < 0.0001), consistent with ROC curve analysis result. NNC model IIand ANN model II exhibited good performance for identifying P450 inhibitors and multi-P450inhibitors (Fig. 4). The global accuracy rateswere 81.3% and 80.0% for identifying P450 inhibitors and 78.7% and 77.0% for identifyingmulti-P450 inhibitors using NNC model II and ANN model II, respectively. Chi-squared testsindicated better performance of NNC model II for identifying P450 inhibitors(p = 0.041) and multi-P450 inhibitors (p < 0.0001)(Table S9). Externalvalidation using 1,919 P450 inhibitors suggested the effectiveness of the above fourmodels (Table S8). Inparticular, NNC model II showed the highest accuracy of 92.1%. Furthermore, we comparedthe efficacies of NNC model II and ANN model II in identifying literature-reported MBIsthat irreversibly inhibit P450s(Table S10). Although the two models did not show different predictions for theMBIs (Chi-squared test, p = 0.41), NNC model II performed better bysuccessfully identifying 126 of the 145 MBIs, whereas ANN model II recognized 121 of the145 MBIs. 10.7717/peerj.1524/fig-4Figure 4Ten-Fold cross-validation of NNC model II and ANN model II.


Artificial neural network cascade identifies multi-P450 inhibitors in natural compounds.

Li Z, Li Y, Sun L, Tang Y, Liu L, Zhu W - PeerJ (2015)

Ten-Fold cross-validation of NNC model II and ANN model II.(A) The AUROCs are 0.864 and 0.842 for discrimination between P450 inhibitors(n = 1–5) and P450 non-inhibitors (n = 0) usingNNC model II and ANN model II, respectively. (B) The AUROCs are 0.845 and 0.822 foridentification of non-multi-P450 inhibitors (n = 0–2) andmulti-P450 inhibitors (n = 3–5) using the two models,respectively.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4696407&req=5

fig-4: Ten-Fold cross-validation of NNC model II and ANN model II.(A) The AUROCs are 0.864 and 0.842 for discrimination between P450 inhibitors(n = 1–5) and P450 non-inhibitors (n = 0) usingNNC model II and ANN model II, respectively. (B) The AUROCs are 0.845 and 0.822 foridentification of non-multi-P450 inhibitors (n = 0–2) andmulti-P450 inhibitors (n = 3–5) using the two models,respectively.
Mentions: The holdout cross-validation method was used for internal validation of each ANN submodelin the two NNC models and the two ANN models. Similar values ofRTe and RTr guaranteedsatisfactory generalizability of the constructed models (Figs. 2 and 3). A set of 2,716 compoundswith complete in vitro P450 inhibition data was applied to test NNC modelI and ANN model I for method validation. The PIS values exported from the two models weresignificantly positively corrected with the normalized number of inhibited P450 isoforms(Spearman’s rho = 0.613 and 0.587 for NNC model I and ANN model I, respectively,p < 0.0001). For NNC model II and ANN model II, the 10-foldcross-hold method was used for internal validation. Significant correlations between thePIS scores and the normalized number of inhibited P450 isoforms were observed for bothmodels (Spearman’s rho = 0.686 and 0.645 for NNC model II and ANN model II, respectively,p < 0.0001), consistent with ROC curve analysis result. NNC model IIand ANN model II exhibited good performance for identifying P450 inhibitors and multi-P450inhibitors (Fig. 4). The global accuracy rateswere 81.3% and 80.0% for identifying P450 inhibitors and 78.7% and 77.0% for identifyingmulti-P450 inhibitors using NNC model II and ANN model II, respectively. Chi-squared testsindicated better performance of NNC model II for identifying P450 inhibitors(p = 0.041) and multi-P450 inhibitors (p < 0.0001)(Table S9). Externalvalidation using 1,919 P450 inhibitors suggested the effectiveness of the above fourmodels (Table S8). Inparticular, NNC model II showed the highest accuracy of 92.1%. Furthermore, we comparedthe efficacies of NNC model II and ANN model II in identifying literature-reported MBIsthat irreversibly inhibit P450s(Table S10). Although the two models did not show different predictions for theMBIs (Chi-squared test, p = 0.41), NNC model II performed better bysuccessfully identifying 126 of the 145 MBIs, whereas ANN model II recognized 121 of the145 MBIs. 10.7717/peerj.1524/fig-4Figure 4Ten-Fold cross-validation of NNC model II and ANN model II.

Bottom Line: Thus, multi-P450 inhibition leads to greater drug-drug interaction risk than specific P450 inhibition.The results indicate significant positive correlation between the PIS values and the number of inhibited P450 isoforms (Spearman's ρ = 0.684, p < 0.0001).Furthermore, chemical similarity calculations suggested that the prevailing parent structures of natural multi-P450 inhibitors were alkaloids.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Pharmacy Administration, Harbin Medical University , Harbin , China.

ABSTRACT
Substantial evidence has shown that most exogenous substances are metabolized by multiple cytochrome P450 (P450) enzymes instead of by merely one P450 isoform. Thus, multi-P450 inhibition leads to greater drug-drug interaction risk than specific P450 inhibition. Herein, we innovatively established an artificial neural network cascade (NNC) model composed of 23 cascaded networks in a ladder-like framework to identify potential multi-P450 inhibitors among natural compounds by integrating 12 molecular descriptors into a P450 inhibition score (PIS). Experimental data reporting in vitro inhibition of five P450 isoforms (CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4) were obtained for 8,148 compounds from the Cytochrome P450 Inhibitors Database (CPID). The results indicate significant positive correlation between the PIS values and the number of inhibited P450 isoforms (Spearman's ρ = 0.684, p < 0.0001). Thus, a higher PIS indicates a greater possibility for a chemical to inhibit the enzyme activity of at least three P450 isoforms. Ten-fold cross-validation of the NNC model suggested an accuracy of 78.7% for identifying whether a compound is a multi-P450 inhibitor or not. Using our NNC model, 22.2% of the approximately 160,000 natural compounds in TCM Database@Taiwan were identified as potential multi-P450 inhibitors. Furthermore, chemical similarity calculations suggested that the prevailing parent structures of natural multi-P450 inhibitors were alkaloids. Our findings show that dissection of chemical structure contributes to confident identification of natural multi-P450 inhibitors and provides a feasible method for virtually evaluating multi-P450 inhibition risk for a known structure.

No MeSH data available.


Related in: MedlinePlus