Towards Deep Neural Network Models for the Prediction of the Blood–Brain Barrier Permeability for Diverse Organic Compounds

Radchenko, Eugene V.; Dyabina, Alina S; Palyulin, Vladimir A.

doi:10.3390/molecules25245901

Cited by 31 publications

(27 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the best of our knowledge, AlphaQ may be viewed as the first reliable 3D-QSAR model for predicting the LogBB data of structurally diverse molecules with no common scaffold. In terms of the predictive capability given by the R 2 test value, AlphaQ outperforms quantum mechanical solvation models [48] as well as 2D-QSAR models involving the deep neural network algorithm [49]. Furthermore, AlphaQ has a computational advantage over atomistic statistical simulations and high-level quantum mechanical calculations in the context that a highly predictive model can be derived in a straightforward way using a moderate amount of experimental data.…”

Section: Resultsmentioning

confidence: 99%

Quantum Artificial Neural Network Approach to Derive a Highly Predictive 3D-QSAR Model for Blood–Brain Barrier Passage

Kim

You

Han

et al. 2021

IJMS

View full text Add to dashboard Cite

A successful passage of the blood–brain barrier (BBB) is an essential prerequisite for the drug molecules designed to act on the central nervous system. The logarithm of blood–brain partitioning (LogBB) has served as an effective index of molecular BBB permeability. Using the three-dimensional (3D) distribution of the molecular electrostatic potential (ESP) as the numerical descriptor, a quantitative structure-activity relationship (QSAR) model termed AlphaQ was derived to predict the molecular LogBB values. To obtain the optimal atomic coordinates of the molecules under investigation, the pairwise 3D structural alignments were conducted in such a way to maximize the quantum mechanical cross correlation between the template and a target molecule. This alignment method has the advantage over the conventional atom-by-atom matching protocol in that the structurally diverse molecules can be analyzed as rigorously as the chemical derivatives with the same scaffold. The inaccuracy problem in the 3D structural alignment was alleviated in a large part by categorizing the molecules into the eight subsets according to the molecular weight. By applying the artificial neural network algorithm to associate the fully quantum mechanical ESP descriptors with the extensive experimental LogBB data, a highly predictive 3D-QSAR model was derived for each molecular subset with a squared correlation coefficient larger than 0.8. Due to the simplicity in model building and the high predictability, AlphaQ is anticipated to serve as an effective computational screening tool for molecular BBB permeability.

show abstract

Section: Resultsmentioning

confidence: 99%

Quantum Artificial Neural Network Approach to Derive a Highly Predictive 3D-QSAR Model for Blood–Brain Barrier Passage

Kim

You

Han

et al. 2021

IJMS

View full text Add to dashboard Cite

show abstract

“…Lipophilicity (LogP ow ) and aqueous solubility (pS) were estimated by the ALogPS 3.0 neural network model implemented in the OCHEM platform [ 116 ]. Human intestinal absorption (HIA) [ 117 ], blood–brain barrier distribution/permeability (LogBB) [ 118 , 119 ], and hERG-mediated cardiac toxicity risk (channel affinity p K i and inhibitory activity pIC 50 ) [ 120 ] were estimated using the integrated online service for the prediction of ADMET properties [ 121 ]. This service implements predictive QSAR models based on accurate and representative training sets, fragmental descriptors, and artificial neural networks.…”

Section: Methodsmentioning

confidence: 99%

Bis-Amiridines as Acetylcholinesterase and Butyrylcholinesterase Inhibitors: N-Functionalization Determines the Multitarget Anti-Alzheimer’s Activity Profile

et al. 2022

Self Cite

View full text Add to dashboard Cite

Using two ways of functionalizing amiridine—acylation with chloroacetic acid chloride and reaction with thiophosgene—we have synthesized new homobivalent bis-amiridines joined by two different spacers—bis-N-acyl-alkylene (3) and bis-N-thiourea-alkylene (5) —as potential multifunctional agents for the treatment of Alzheimer’s disease (AD). All compounds exhibited high inhibitory activity against acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) with selectivity for BChE. These new agents displayed negligible carboxylesterase inhibition, suggesting a probable lack of untoward drug–drug interactions arising from hydrolytic biotransformation. Compounds 3 with bis-N-acyl-alkylene spacers were more potent inhibitors of both cholinesterases compared to compounds 5 and the parent amiridine. The lead compounds 3a–c exhibited an IC50(AChE) = 2.9–1.4 µM, IC50(BChE) = 0.13–0.067 µM, and 14–18% propidium displacement at 20 mM. Kinetic studies of compounds 3a and 5d indicated mixed-type reversible inhibition. Molecular docking revealed favorable poses in both catalytic and peripheral AChE sites. Propidium displacement from the peripheral site by the hybrids suggests their potential to hinder AChE-assisted Aβ42 aggregation. Conjugates 3 had no effect on Aβ42 self-aggregation, whereas compounds 5c–e (m = 4, 5, 6) showed mild (13–17%) inhibition. The greatest difference between conjugates 3 and 5 was their antioxidant activity. Bis-amiridines 3 with N-acylalkylene spacers were nearly inactive in ABTS and FRAP tests, whereas compounds 5 with thiourea in the spacers demonstrated high antioxidant activity, especially in the ABTS test (TEAC = 1.2–2.1), in agreement with their significantly lower HOMO-LUMO gap values. Calculated ADMET parameters for all conjugates predicted favorable blood–brain barrier permeability and intestinal absorption, as well as a low propensity for cardiac toxicity. Thus, it was possible to obtain amiridine derivatives whose potencies against AChE and BChE equaled (5) or exceeded (3) that of the parent compound, amiridine. Overall, based on their expanded and balanced pharmacological profiles, conjugates 5c-e appear promising for future optimization and development as multitarget anti-AD agents.

show abstract

“…The lipophilicity (LogPow) and aqueous solubility (pSaq) were estimated using the ALogPS 3.0 neural network model implemented in the OCHEM platform. [39] Human intestinal absorption, [40] blood-brain barrier permeability, [41,42] and hERG-mediated cardiac toxicity risk (channel affinity pK i and inhibitory activity pIC 50 ) [43] were estimated using the integrated online service for prediction of ADMET properties (ADMET Prediction Service). [44] This server implements predictive QSAR models based on accurate and representative training sets, fragmental descriptors, and artificial neural networks.…”

Section: Prediction Of Admet Physicochemical and Pains Profilesmentioning

confidence: 99%

Novel substituted 5‐methyl‐4‐acylaminoisoxazoles as antimitotic agents: Evaluation of selectivity to LNCaP cancer cells

Sadovnikov

Vasilenko

Gracheva

et al. 2022

Archiv der Pharmazie

Self Cite

View full text Add to dashboard Cite

A series of novel antimitotic agents was designed using the replacement of heterocyclic cores in two tubulin‐targeting lead molecules with the acylated 4‐aminoisoxazole moiety. Target compounds were synthesized via heterocyclization of β‐aryl‐substituted vinylketones by tert‐butyl nitrite in the presence of water as a key step. 4‐Methyl‐N‐[5‐methyl‐3‐(3,4,5‐trimethoxyphenyl)isoxazol‐4‐yl]benzamide (1aa) was found to stimulate partial depolymerization of microtubules of human lung carcinoma A549 cells at a high concentration of 100 µM and to totally inhibit cell growth (IC50 = 0.99 µM) and cell viability (IC50 = 0.271 µM) in the nanomolar to submicromolar concentration range. These data provide evidence of the multitarget profile of the cytotoxic action of compound 1aa. The SAR study demonstrated that the 3,4,5‐trimethoxyphenyl residue is the key structural parameter determining the efficiency both towards tubulin and other molecular targets. The cytotoxicity of 3‐methyl‐N‐[5‐methyl‐3‐(3,4,5‐trimethoxyphenyl)isoxazol‐4‐yl]benzamide (1ab) to the androgen‐sensitive human prostate adenocarcinoma cancer cell line LNCaP (IC50 = 0.301 µM) was approximately one order of magnitude higher than that to the conditionally normal cells lines WI‐26 VA4 (IC50 = 2.26 µM) and human umbilical vein endothelial cells (IC50 = 5.58 µM) and significantly higher than that to primary fibroblasts (IC50 > 75 µM).

show abstract

Towards Deep Neural Network Models for the Prediction of the Blood–Brain Barrier Permeability for Diverse Organic Compounds

Cited by 31 publications

References 79 publications

Quantum Artificial Neural Network Approach to Derive a Highly Predictive 3D-QSAR Model for Blood–Brain Barrier Passage

Quantum Artificial Neural Network Approach to Derive a Highly Predictive 3D-QSAR Model for Blood–Brain Barrier Passage

Bis-Amiridines as Acetylcholinesterase and Butyrylcholinesterase Inhibitors: N-Functionalization Determines the Multitarget Anti-Alzheimer’s Activity Profile

Novel substituted 5‐methyl‐4‐acylaminoisoxazoles as antimitotic agents: Evaluation of selectivity to LNCaP cancer cells

Contact Info

Product

Resources

About