Chapter 6. Representing Chemical Structures in Databases for Drug Design

2011 ◽  
pp. 164-191 ◽  
Author(s):  
John M. Barnard ◽  
Peter W. Kenny ◽  
Paul N. Wallace
Keyword(s):  
Drug Design ◽  
Chemical Structures

Novel Scopoletin Derivatives Kill Cancer Cells by Inducing Mitochondrial Depolarization and Apoptosis

2020 ◽  
Vol 21 ◽  
Author(s):  
Zhixian Shi ◽  
Li Chen ◽  
Jianbo Sun
Keyword(s):  
Drug Design ◽  
Anticancer Activity ◽  
Cancer Cells ◽  
Antiproliferative Activity ◽  
Mitochondrial Depolarization ◽  
Irreversible Inhibitor ◽  
Apoptotic Pathway ◽  
Normal Cells ◽  
Chemical Structures

Background: Natural products and their molecular frameworks have been explored as invaluable sources of inspiration for drug design by means of structural modification, computer aided drug design, and so on. Scopoletin extracting from multiple herbs exhibits potential anticancer activity in vitro and vivo without toxicity towards normal cells. Objective: To obtain new scopoletin derivatives with enhanced anticancer activity, we performed the chemical structure modification and researched the mechanism of anti-tumor activity. Methods: In this study, we take regard scopoletin as lead compound, designed and synthesized a series of scopoletin derivatives via introducing different heterocyclic fragments, and their chemical structures were characterized by NMR spectra (1H NMR and 13C NMR) and HRMS(ESI). The antiproliferative activity of target compounds in four cancer cell lines (MDA-MB-231, MCF-7, HepG2, and A549) were determined by the MTT assay. Compound 11b was treated with Ac-cys under different reaction condition to explore the thiol addition activity of it. The Annexin V/PI and JC-1 staining assay were performed to investigate the anti-tumor mechanism of 11b. Results: Novel compounds 8a-h and 11a-h derivatives of scopoletin were synthesized. Most of target compounds exhibited enhanced antiproliferative activity against different cancer cells and reduced toxicity towards normal cells. In particular, 11b displayed the optimal antitumor ability against breast cancer MDA-MB-231 cells with an IC50 value of 4.46 μM. 11b also cannot react with Ac-cys under the experimental condition. When treated with 11b for 24 h, the total apoptotic cells increased from 10.8% to 79.3%. Besides, 11b induced the depolarization of mitochondrial membrane potential. Conclusion: 11b was more active than other derivatives, indicating that the introduction of thiophene fragment was beneficial for the enhancement of antitumor effect, and it was also not an irreversible inhibitor basing on the result that the α, β-unsaturated ketones of 11b cannot undergo Michael addition reactions with Ac-cys. Furthermore, studies on the pharmacological mechanism showed that 11b induced the mitochondrial depolarization and apoptosis, which indicated 11b killed cancer cells via mitochondrial apoptotic pathway. Therefore, an in-depth research and structure optimization of this compound is warranted.

scholarly journals DeepCDR: a hybrid graph convolutional network for predicting cancer drug response

2020 ◽  
Author(s):  
Qiao Liu ◽  
Zhiqiang Hu ◽  
Rui Jiang ◽  
Mu Zhou
Keyword(s):  
Drug Design ◽  
Cancer Biology ◽  
Drug Response ◽  
Drug Efficacy ◽  
Supplementary Information ◽  
Cancer Drug ◽  
Convolutional Network ◽  
Chemical Structures ◽  
Anti Cancer ◽  
Cancer Types

AbstractMotivationAccurate prediction of cancer drug response (CDR) is challenging due to the uncertainty of drug efficacy and heterogeneity of cancer patients. Strong evidences have implicated the high dependence of CDR on tumor genomic and transcriptomic profiles of individual patients. Precise identification of CDR is crucial in both guiding anti-cancer drug design and understanding cancer biology.ResultsIn this study, we present DeepCDR which integrates multi-omics profiles of cancer cells and explores intrinsic chemical structures of drugs for predicting cancer drug response. Specifically, DeepCDR is a hybrid graph convolutional network consisting of a uniform graph convolutional network (UGCN) and multiple subnetworks. Unlike prior studies modeling hand-crafted features of drugs, DeepCDR automatically learns the latent representation of topological structures among atoms and bonds of drugs. Extensive experiments showed that DeepCDR outperformed state-of-the-art methods in both classification and regression settings under various data settings. We also evaluated the contribution of different types of omics profiles for assessing drug response. Furthermore, we provided an exploratory strategy for identifying potential cancer-associated genes concerning specific cancer types. Our results highlighted the predictive power of DeepCDR and its potential translational value in guiding disease-specific drug design.AvailabilityDeepCDR is freely available at https://github.com/kimmo1019/[email protected]; [email protected] informationSupplementary data are available at Bioinformatics online.

scholarly journals Retro Drug Design: From Target Properties to Molecular Structures

2021 ◽  
Author(s):  
Yuhong Wang ◽  
Sam Michael ◽  
Ruili Huang ◽  
Jinghua Zhao ◽  
Katlin Recabo ◽  
...  
Keyword(s):  
Drug Discovery ◽  
Drug Design ◽  
Molecular Descriptor ◽  
Pharmaceutical Research ◽  
Monte Carlo Sampling ◽  
Molecular Structures ◽  
Descriptor System ◽  
Chemical Structures ◽  
Drug Molecules ◽  
Property Space

To generate drug molecules of desired properties with computational methods is the holy grail in pharmaceutical research. Here we describe an AI strategy, retro drug design, or RDD, to generate novel small molecule drugs from scratch to meet predefined requirements, including but not limited to biological activity against a drug target, and optimal range of physicochemical and ADMET properties. Traditional predictive models were first trained over experimental data for the target properties, using an atom typing based molecular descriptor system, ATP. Monte Carlo sampling algorithm was then utilized to find the solutions in the ATP space defined by the target properties, and the deep learning model of Seq2Seq was employed to decode molecular structures from the solutions. To test feasibility of the algorithm, we challenged RDD to generate novel drugs that can activate μ opioid receptor (MOR) and penetrate blood brain barrier (BBB). Starting from vectors of random numbers, RDD generated 180,000 chemical structures, of which 78% were chemically valid. About 42,000 (31%) of the valid structures fell into the property space defined by MOR activity and BBB permeability. Out of the 42,000 structures, only 267 chemicals were commercially available, indicating a high extent of novelty of the AI-generated compounds. We purchased and assayed 96 compounds, and 25 of which were found to be MOR agonists. These compounds also have excellent BBB scores. The results presented in this paper illustrate that RDD has potential to revolutionize the current drug discovery process and create novel structures with multiple desired properties, including biological functions and ADMET properties. Availability of an AI-enabled fast track in drug discovery is essential to cope with emergent public health threat, such as pandemic of COVID-19.

scholarly journals Sanitize It Yourself: human-based sanitization checker against machine-generated chemical structures

2021 ◽  
Author(s):  
Naruki Yoshikawa ◽  
Kentaro Rikimaru ◽  
Kazuki Yamamoto
Keyword(s):  
Deep Learning ◽  
Drug Design ◽  
Web Application ◽  
Chemical Space ◽  
Computer Aided Drug Design ◽  
Chemical Structures ◽  
Structural Formulas ◽  
Computer Aided ◽  
Novel Structures

Many computer-aided drug design (CADD) methods using deep learning have recently been proposed to explore the chemical space toward novel scaffolds efficiently. However, there is a tradeoff between the ease of generating novel structures and the chemical feasibility of structural formulas. To overcome the limitations of computational filtering, we have implemented a web application that allows easy compound sanitization by humans. The application is available at https://sanitizer.chemical.space/.

scholarly journals Computing Zagreb and Randić Indices of PEG-Cored Dendrimers Used for Drug and Gene Delivery

2020 ◽  
Vol 2020 ◽  
pp. 1-7
Author(s):  
Thayamathy Pio Jude ◽  
Elango Panchadcharam ◽  
Koneswaran Masilamani
Keyword(s):  
Gene Delivery ◽  
Drug Design ◽  
Topological Indices ◽  
Randić Index ◽  
Molecular Topology ◽  
Design And Development ◽  
Randic Index ◽  
Zagreb Indices ◽  
Chemical Structures ◽  
Drug And Gene Delivery

Zagreb and Randić indices are the most commonly used degree-based topological indices in the study of drug design and development. In molecular topology, M-polynomials are also used to calculate the degree-based topological indices of chemical structures. In this paper, we derive the M-polynomials for the PEG-cored PAMAM, carbosilane, and poly (lysine) dendrimers and calculate their first, second, and second modified Zagreb indices and the Randić index.

OpenChem: A Deep Learning Toolkit for Computational Chemistry and Drug Design

2020 ◽  
Author(s):  
Mariya Popova ◽  
Boris Ginsburg ◽  
Alexander Tropsha ◽  
Olexandr Isayev
Keyword(s):  
Deep Learning ◽  
Drug Design ◽  
Computational Chemistry ◽  
Language Processing ◽  
Nearest Neighbor ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Materials Informatics ◽  
Learning Models ◽  
Chemical Structures

Deep learning models have demonstrated outstanding results in many data-rich areas of research, such as computer vision and natural language processing. Currently, there is a rise of deep learning in computational chemistry and materials informatics, where deep learning could be effectively applied in modeling the relationship between chemical structures and their properties. With the immense growth of chemical and materials data, deep learning models can begin to outperform conventional machine learning techniques such as random forest, support vector machines, nearest neighbor, etc. Herein, we introduce OpenChem, a PyTorch-based deep learning toolkit for computational chemistry and drug design. OpenChem offers easy and fast model development, modular software design, and several data preprocessing modules. It is freely available via the GitHub repository.

OpenChem: A Deep Learning Toolkit for Computational Chemistry and Drug Design

2020 ◽  
Author(s):  
Mariya Popova ◽  
Boris Ginsburg ◽  
Alexander Tropsha ◽  
Olexandr Isayev
Keyword(s):  
Deep Learning ◽  
Drug Design ◽  
Computational Chemistry ◽  
Language Processing ◽  
Nearest Neighbor ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Materials Informatics ◽  
Learning Models ◽  
Chemical Structures

Deep learning models have demonstrated outstanding results in many data-rich areas of research, such as computer vision and natural language processing. Currently, there is a rise of deep learning in computational chemistry and materials informatics, where deep learning could be effectively applied in modeling the relationship between chemical structures and their properties. With the immense growth of chemical and materials data, deep learning models can begin to outperform conventional machine learning techniques such as random forest, support vector machines, nearest neighbor, etc. Herein, we introduce OpenChem, a PyTorch-based deep learning toolkit for computational chemistry and drug design. OpenChem offers easy and fast model development, modular software design, and several data preprocessing modules. It is freely available via the GitHub repository.

Computer-Aided Drug Design Applied to Secondary Metabolites as Anticancer Agents

2020 ◽  
Vol 20 (19) ◽  
pp. 1677-1703
Author(s):  
Rodrigo Santos Aquino de Araújo ◽  
Edeildo Ferreira da Silva-Junior ◽  
Thiago Mendonça de Aquino ◽  
Marcus Tullius Scotti ◽  
Hamilton M. Ishiki ◽  
...  
Keyword(s):  
Drug Design ◽  
Anticancer Agents ◽  
Cost Reduction ◽  
Computer Aided Drug Design ◽  
Structure Based Drug Design ◽  
Chemical Structures ◽  
Computer Aided ◽  
Low Costs ◽  
New Compounds

: Computer-Aided Drug Design (CADD) techniques have garnered a great deal of attention in academia and industry because of their great versatility, low costs, possibilities of cost reduction in in vitro screening and in the development of synthetic steps; these techniques are compared with highthroughput screening, in particular for candidate drugs. The secondary metabolism of plants and other organisms provide substantial amounts of new chemical structures, many of which have numerous biological and pharmacological properties for virtually every existing disease, including cancer. In oncology, compounds such as vimblastine, vincristine, taxol, podophyllotoxin, captothecin and cytarabine are examples of how important natural products enhance the cancer-fighting therapeutic arsenal. : In this context, this review presents an update of Ligand-Based Drug Design and Structure-Based Drug Design techniques applied to flavonoids, alkaloids and coumarins in the search of new compounds or fragments that can be used in oncology. : A systematical search using various databases was performed. The search was limited to articles published in the last 10 years. : The great diversity of chemical structures (coumarin, flavonoids and alkaloids) with cancer properties, associated with infinite synthetic possibilities for obtaining analogous compounds, creates a huge chemical environment with potential to be explored, and creates a major difficulty, for screening studies to select compounds with more promising activity for a selected target. CADD techniques appear to be the least expensive and most efficient alternatives to perform virtual screening studies, aiming to selected compounds with better activity profiles and better “drugability”.

Sign in / Sign up

Export Citation Format

Share Document