Predicting the Drug Clearance Pathway with Structural Descriptors

Kaboudi, Navid; Shayanfar, Ali

doi:10.1007/s13318-021-00748-3

Cited by 7 publications

(7 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The calculated AUC of the predictive models was in the range of 0.776–0.870, and the calculated accuracy (ACC) was in the range of 0.72–0.77. 10 Lombardo et al constructed a predictive model using MDs with an ACC of 0.84, although the compounds used differed from those used in the present investigation. 34 In the present study, we used a data set of 636 compounds created by Kaboudi and Shayanfar based on a report by Lombardo, which are equally divided chemical space ( Figure 1 ), to investigate the predictive models combining the DeepSnap-DL and MD-based methods.…”

Section: Discussionmentioning

confidence: 99%

“…However, predictions of the CL pathway using only the structural information of compounds have been reported by Kaboudi and Shayanfar and by Lombardo et al , Kaboudi and Shayanfar divided compounds randomly and then constructed predictive models using MDs. The calculated AUC of the predictive models was in the range of 0.776–0.870, and the calculated accuracy (ACC) was in the range of 0.72–0.77 . Lombardo et al constructed a predictive model using MDs with an ACC of 0.84, although the compounds used differed from those used in the present investigation .…”

Section: Discussionmentioning

confidence: 99%

“…For the prediction of CL pathways, the extended clearance classification system using experimental values of membrane permeability has been reported, although experimental values are needed to perform this prediction . However, predictions of the CL pathway using only the structural information of compounds have been reported by Kaboudi and Shayanfar and by Lombardo et al , Kaboudi and Shayanfar divided compounds randomly and then constructed predictive models using MDs. The calculated AUC of the predictive models was in the range of 0.776–0.870, and the calculated accuracy (ACC) was in the range of 0.72–0.77 .…”

Section: Discussionmentioning

confidence: 99%

“…To reduce the use of animal experiments and developmental expenses, predictive models based on quantitative structure–activity relationship (QSAR) analysis have been actively used in recent years at the drug discovery stage. − QSAR analysis is an approach based on features of compounds such as molecular descriptors (MDs) and fingerprints, and algorithms such as support vector machine, random forest, artificial neural network, k-nearest neighbor, XGBoost, and deep learning (DL) have been applied. − At the drug discovery stage, QSAR analysis is used to predict pharmacological activity, pharmacokinetic parameters, and toxicity parameters. Several predictive models based on QSAR analysis were previously reported for parameters associated with toxicity, such as inhibition of the human ether-a-go-go related gene and 50% lethal dose (LD 50 ), and parameters associated with pharmacokinetics, such as metabolic stability, protein binding, distribution to blood cells, membrane permeability, clearance (CL), and the CL pathway. − Therefore, these predictive models are used by various pharmaceutical companies to perform virtual screening, narrow hit compounds, and prioritize various experiments. − , Thus, predictive models have been reported for many toxicity parameters and pharmacokinetic parameters. − , However, for some targets, the predictive performance of these models is apparently insufficient. As an approach to improve the predictive performance for targets such as LD 50 , a consensus model that combines predictive models constructed by 35 different organizations was reported .…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Mamada,

Takahashi,

Ogino

et al. 2023

ACS Omega

View full text Add to dashboard Cite

Various toxicity and pharmacokinetic evaluations as screening experiments are needed at the drug discovery stage. Currently, to reduce the use of animal experiments and developmental expenses, the development of high-performance predictive models based on quantitative structure–activity relationship analysis is desired. From these evaluation targets, we selected 50% lethal dose (LD50), blood–brain barrier penetration (BBBP), and the clearance (CL) pathway for this investigation and constructed predictive models for each target using 636–11,886 compounds. First, we constructed predictive models using the DeepSnap-deep learning (DL) method and images of compounds as features. The calculated area under the curve (AUC) and balanced accuracy (BAC) were, respectively, 0.887 and 0.818 for LD50, 0.893 and 0.824 for BBBP, and 0.883 and 0.763 for the CL pathway. Next, molecular descriptors (MDs) of compounds were calculated using Molecular Operating Environment, alvaDesc, and ADMET Predictor to construct predictive models using the MD-based method. Using these MDs, we constructed predictive models using DataRobot. The calculated AUC and BAC were, respectively, 0.931 and 0.805 for LD50, 0.919 and 0.849 for BBBP, and 0.900 and 0.807 for the CL pathway. In this investigation, we constructed predictive models combining the DeepSnap-DL and MD-based methods. In ensemble models using the mean predictive probability of the DeepSnap-DL and MD-based methods, the calculated AUC and BAC were, respectively, 0.942 and 0.842 for LD50, 0.936 and 0.853 for BBBP, and 0.908 and 0.832 for the CL pathway, with improved predictive performance observed for all variables compared with either single method alone. Moreover, in consensus models that adopted only compounds for which the results of the two methods agreed, the calculated BAC for LD50, BBBP, and the CL pathway were 0.916, 0.918, and 0.847, respectively, indicating higher predictive performance than the ensemble models for all three variables. The predictive models combining the DeepSnap-DL and MD-based methods displayed high predictive performance for LD50, BBBP, and the CL pathway. Therefore, the application of this approach to prediction targets in various drug discovery screenings is expected to accelerate drug discovery.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Mamada,

Takahashi,

Ogino

et al. 2023

ACS Omega

View full text Add to dashboard Cite

show abstract

“…Besides the TPSA and LogD, other properties such as the net charge, number of rotatable bonds, and number of hydrogen bonds showed close relationships with renal and hepatobiliary clearance (Figure S6, Supporting Information). [18][19][20][21] In general, the influence pattern of polar descriptors on hepatobiliary clearance is less clear due to the limited number of fluorophores getting stuck in the liver, with the largest percentage stuck being zwitterionic (Figure S6a,b, Supporting Information). On the other hand, a significant number of fluorophores with a þ1 or þ2 charge were stuck in the kidney, whereas zwitterionic fluorophores (net charge of 0) had a smaller percentage stuck in the kidneys (Figure S6a, Supporting Information).…”

Section: Polar Descriptorsmentioning

confidence: 99%

Physicochemical Descriptors in Biodistribution and Clearance of Contrast Agents

Ahn,

Sanchez-Langeling,

Lee

et al. 2023

Advanced Photonics Research

View full text Add to dashboard Cite

Hepatobiliary and renal excretions are the two major in vivo elimination pathways, of which prediction is particularly important for the optimization of systemic and/or target‐site exposure of new near‐infrared (NIR) fluorescent contrast agents. To characterize the physicochemical descriptors associated with each clearance route, a compound library containing more than 760 NIR fluorophores is analyzed and systematically reviewed in this study. Although there are a host of biochemical and physiological mechanisms that ultimately determine the in vivo fate of any given molecule, the physicochemical properties of contrast agents including molecular weight, hydrophobicity, topological polar surface area, molecular charges, and hydrogen/rotatable bonds are often indicative of each clearance route. The approach developed is anticipated to be useful in early lead identification studies when selecting NIR fluorescent contrast agents from large database screenings. It may also be applied to prioritize synthetically feasible chemical modifications during lead compound optimization.

show abstract