Pub Date : 2019-05-11DOI: 10.1504/IJCBDD.2019.10021269
Saurav Mallik, Zhongming Zhao
Biomarker discovery from complex biomedical data has become an important topic to unveil the significant new disease signals for disease diagnosis and treatment during past two decades. The earlier methods were proposed on a single genomic profile, and most of them utilize a single minimum support/confidence/lift cutoff. To overcome these shortcomings, here, we developed a framework for identifying complex markers using shortest distance based rule mining from the tri-omics profiles (gene expression, methylation and protein-protein interaction). We applied our method to a high-grade soft-tissue sarcomas multi-omics dataset. The novel markers were {GRB2-, STAT3-}('-' and '+' denote decreased and increased gene activities, respectively), {STAT3+, TP53-, MAPK3+} and {STAT3+, FYN+, MAPK3+}. We showed the superiority of our method vs. others, as it generates fewer rules and lower mean of the shortest distance than others. Moreover, our method is useful to extract complex markers from tri-omics profiles for the complex disease.
{"title":"Distance based knowledge retrieval through rule mining for complex biomarker recognition from tri-omics profiles","authors":"Saurav Mallik, Zhongming Zhao","doi":"10.1504/IJCBDD.2019.10021269","DOIUrl":"https://doi.org/10.1504/IJCBDD.2019.10021269","url":null,"abstract":"Biomarker discovery from complex biomedical data has become an important topic to unveil the significant new disease signals for disease diagnosis and treatment during past two decades. The earlier methods were proposed on a single genomic profile, and most of them utilize a single minimum support/confidence/lift cutoff. To overcome these shortcomings, here, we developed a framework for identifying complex markers using shortest distance based rule mining from the tri-omics profiles (gene expression, methylation and protein-protein interaction). We applied our method to a high-grade soft-tissue sarcomas multi-omics dataset. The novel markers were {GRB2-, STAT3-}('-' and '+' denote decreased and increased gene activities, respectively), {STAT3+, TP53-, MAPK3+} and {STAT3+, FYN+, MAPK3+}. We showed the superiority of our method vs. others, as it generates fewer rules and lower mean of the shortest distance than others. Moreover, our method is useful to extract complex markers from tri-omics profiles for the complex disease.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"12 1","pages":"105-127"},"PeriodicalIF":0.0,"publicationDate":"2019-05-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87157677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-03-07DOI: 10.1504/IJCBDD.2019.098178
G. Khensous, B. Messabih, Abdallah Chouarfia, B. Maigret
In this paper, we present a molecular docking method to predict the optimal binding pose of a flexible ligand in a flexible protein-binding pocket. For this purpose, a Tabu global search optimization algorithm is used, and the best Tabu solutions are then refined using the Nelder-Mead Simplex local search optimization algorithm. Most docking methods use scoring functions to approximate the binding affinity between the two molecular partners. In our application, the intra-molecular and intermolecular energies are calculated explicitly from a classical molecular mechanics model, which includes polarization terms. The variables of our optimization problem are the ligand positions (Euler angles + translation vector), the ligand and the protein side chains dihedral angles instead of the Cartesian coordinates in order to reduce the problem dimensionality. While the GOLD software (GOLD for Genetic Optimization for Ligand Docking) is usually considered as a standard in molecular docking, our docking approach is illustrated on four protein/ligand complexes for which GOLD failed, suggesting that the proposed method is promising.
在本文中，我们提出了一种分子对接方法来预测柔性配体在柔性蛋白质结合口袋中的最佳结合姿态。为此，使用Tabu全局搜索优化算法，然后使用Nelder-Mead Simplex局部搜索优化算法对最佳Tabu解进行细化。大多数对接方法使用评分函数来近似两个分子伴侣之间的结合亲和力。在我们的应用中，分子内和分子间的能量是由一个经典的分子力学模型明确地计算出来的，其中包括极化项。优化问题的变量为配体位置(欧拉角+平移向量)、配体与蛋白质侧链的二面角，以降低问题的维数。虽然GOLD软件(GOLD for Genetic Optimization for Ligand Docking)通常被认为是分子对接的标准，但我们的对接方法在四个GOLD失败的蛋白质/配体复合物上进行了说明，表明所提出的方法是有前途的。
{"title":"Flexible molecular docking: application of hybrid tabu-simplex optimisation","authors":"G. Khensous, B. Messabih, Abdallah Chouarfia, B. Maigret","doi":"10.1504/IJCBDD.2019.098178","DOIUrl":"https://doi.org/10.1504/IJCBDD.2019.098178","url":null,"abstract":"In this paper, we present a molecular docking method to predict the optimal binding pose of a flexible ligand in a flexible protein-binding pocket. For this purpose, a Tabu global search optimization algorithm is used, and the best Tabu solutions are then refined using the Nelder-Mead Simplex local search optimization algorithm. Most docking methods use scoring functions to approximate the binding affinity between the two molecular partners. In our application, the intra-molecular and intermolecular energies are calculated explicitly from a classical molecular mechanics model, which includes polarization terms. The variables of our optimization problem are the ligand positions (Euler angles + translation vector), the ligand and the protein side chains dihedral angles instead of the Cartesian coordinates in order to reduce the problem dimensionality. While the GOLD software (GOLD for Genetic Optimization for Ligand Docking) is usually considered as a standard in molecular docking, our docking approach is illustrated on four protein/ligand complexes for which GOLD failed, suggesting that the proposed method is promising.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"1 1","pages":"34-53"},"PeriodicalIF":0.0,"publicationDate":"2019-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82254767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-03-07DOI: 10.1504/IJCBDD.2019.098183
Shivananda Kandagalla, S. Shekarappa, Bharath Basavapattana Rudresh, Pavan Gollapalli, M. Hanumanthappa
TGF-β signalling is a key mediator of epithelial to mesenchymal transition (EMT) process and its up-regulation is identified as a hallmark of metastasis. Since TGF-β signalling pathway is known as a key therapeutic target in the treatment of EMT enabled cancer and the study aims at identification of key EMT genes by gene annotation tools and protein interaction network (PIN) to analyse the regulatory dynamics of an interactome. Meanwhile, the potency of curcumin against TGF-β signalling was evaluated by network pharmacology approach. Resultantly, 15 genes were identified as key regulators of TGF-β signalling pathway and seven were shortlisted as leading curcumin targets. Cumulatively, both approaches have justified the role of targets. Thus, curcumin was subjected to molecular docking with targets using AutoDock Vina. Wherein, curcumin has shown significant binding energy with targets EP300 and JUN (-7.1 and -6.4 kcal/mol) respectively indicating the potential anticancer property.
{"title":"Protein interaction network analysis of TGF-β signalling pathway enabled EMT process to anticipate the anticancer activity of curcumin","authors":"Shivananda Kandagalla, S. Shekarappa, Bharath Basavapattana Rudresh, Pavan Gollapalli, M. Hanumanthappa","doi":"10.1504/IJCBDD.2019.098183","DOIUrl":"https://doi.org/10.1504/IJCBDD.2019.098183","url":null,"abstract":"TGF-β signalling is a key mediator of epithelial to mesenchymal transition (EMT) process and its up-regulation is identified as a hallmark of metastasis. Since TGF-β signalling pathway is known as a key therapeutic target in the treatment of EMT enabled cancer and the study aims at identification of key EMT genes by gene annotation tools and protein interaction network (PIN) to analyse the regulatory dynamics of an interactome. Meanwhile, the potency of curcumin against TGF-β signalling was evaluated by network pharmacology approach. Resultantly, 15 genes were identified as key regulators of TGF-β signalling pathway and seven were shortlisted as leading curcumin targets. Cumulatively, both approaches have justified the role of targets. Thus, curcumin was subjected to molecular docking with targets using AutoDock Vina. Wherein, curcumin has shown significant binding energy with targets EP300 and JUN (-7.1 and -6.4 kcal/mol) respectively indicating the potential anticancer property.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"32 1","pages":"54-79"},"PeriodicalIF":0.0,"publicationDate":"2019-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85040203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-03-07DOI: 10.1504/IJCBDD.2019.098177
Pravin Kumar, Shashwati Ghosh Sachan, R. Poddar
Biotransformation of ferulic acid by microorganisms provides a better alternative for production of flavour and fragrance compounds like 4-vinylguaiacol and vanillin. Ferulic acid is transformed to 4-vinylguaiacol using the non-oxidative decarboxylation pathway by ferulic acid decarboxylase (FADase). Here we report, computational mutational analysis of active site of FADase. Site directed mutations (single nucleotide polymorphisms, SNPs) were commenced using in-silico molecular modelling methods. Energy minimisation, dynamic cross-correlation map (DCCM) and principle components analysis (PCA) methods were subsequently applied to validate different conformers (SNPs) of FADase. Substrate ferulic acid was docked with different SNPs. It was observed that, certain amino acids like Tyr21, Trp25, Tyr27 and Glu134 at active sites are responsible for better binding to ferulic acid. Further, mutated form Y27F (Tyr27Phe) of FADase shows a better binding affinity towards ferulic acid than its native form through structure analysis and docking studies.
{"title":"In-silico mutational study of ferulic acid decarboxylase for improvement of substrate binding empathy","authors":"Pravin Kumar, Shashwati Ghosh Sachan, R. Poddar","doi":"10.1504/IJCBDD.2019.098177","DOIUrl":"https://doi.org/10.1504/IJCBDD.2019.098177","url":null,"abstract":"Biotransformation of ferulic acid by microorganisms provides a better alternative for production of flavour and fragrance compounds like 4-vinylguaiacol and vanillin. Ferulic acid is transformed to 4-vinylguaiacol using the non-oxidative decarboxylation pathway by ferulic acid decarboxylase (FADase). Here we report, computational mutational analysis of active site of FADase. Site directed mutations (single nucleotide polymorphisms, SNPs) were commenced using in-silico molecular modelling methods. Energy minimisation, dynamic cross-correlation map (DCCM) and principle components analysis (PCA) methods were subsequently applied to validate different conformers (SNPs) of FADase. Substrate ferulic acid was docked with different SNPs. It was observed that, certain amino acids like Tyr21, Trp25, Tyr27 and Glu134 at active sites are responsible for better binding to ferulic acid. Further, mutated form Y27F (Tyr27Phe) of FADase shows a better binding affinity towards ferulic acid than its native form through structure analysis and docking studies.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"41 1","pages":"16-33"},"PeriodicalIF":0.0,"publicationDate":"2019-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76238915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-03-07DOI: 10.1504/IJCBDD.2019.098180
T. H. Ogunwa
Angelica polymorpha and Beilschmiedia pulverulenta are medicinal plants locally used by people in some parts of Asia and Africa due to their beneficial health effects particularly in the treatment of Alzheimer's disease (AD). The phytoconstituents responsible for such bioactivity have recently been identified in the plants. Herein, in silico approach was used to explore the interaction of such phytochemicals with acetylcholinesterase (AChE) as a validated target in the treatment of AD to provide insights into their precise binding pattern and affinity, order of chemical interaction, inhibitory potential and residues that contribute to the enzyme-phytoconstituent complex stability. With binding affinity ranging from -7.0 kcal/mol to -10.2 kcal/mol and tacrine-comparable orientation, the chemical scaffold of the phytochemicals from both plants displayed deep penetration and fit conveniently into the narrow gorge of AChE. Optimisation of these ligands scaffold might yield new AChE inhibitors with desirable higher efficacy.
{"title":"Interaction studies of Angelica polymorpha and Beilschmiedia pulverulenta phytochemicals with acetylcholinesterase as anti-Alzheimer's disease target","authors":"T. H. Ogunwa","doi":"10.1504/IJCBDD.2019.098180","DOIUrl":"https://doi.org/10.1504/IJCBDD.2019.098180","url":null,"abstract":"Angelica polymorpha and Beilschmiedia pulverulenta are medicinal plants locally used by people in some parts of Asia and Africa due to their beneficial health effects particularly in the treatment of Alzheimer's disease (AD). The phytoconstituents responsible for such bioactivity have recently been identified in the plants. Herein, in silico approach was used to explore the interaction of such phytochemicals with acetylcholinesterase (AChE) as a validated target in the treatment of AD to provide insights into their precise binding pattern and affinity, order of chemical interaction, inhibitory potential and residues that contribute to the enzyme-phytoconstituent complex stability. With binding affinity ranging from -7.0 kcal/mol to -10.2 kcal/mol and tacrine-comparable orientation, the chemical scaffold of the phytochemicals from both plants displayed deep penetration and fit conveniently into the narrow gorge of AChE. Optimisation of these ligands scaffold might yield new AChE inhibitors with desirable higher efficacy.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"1 1","pages":"80-99"},"PeriodicalIF":0.0,"publicationDate":"2019-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88314887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-03-07DOI: 10.1504/IJCBDD.2019.098175
P. Swaminathan, L. Saleena
Malaria still remains one of the challenging public health issue infecting about 300-500 millions of people. The most serious and fatal malarial infections are caused by Plasmodium falciparum which has developed resistance to commonly employed therapeutics. Hence the need to develop a novel anti-malarial drug targeting Dihydroorotate dehydrogenase (DHODH), an enzyme involved in parasite growth. DHODH is present in both humans and Plasmodium falciparum. Sequence analysis and structure comparison of DHODH of both Human and Plasmodium falciparum reveals variations among them, thereby providing a chance to design a specific inhibitor. Virtual screening of existing anti-malarial drugs acting on DHODH is performed from Pubchem and BindingDB databases. Pharmacophore mapping was done for the top 20 virtual screening compounds using hip hop algorithm. The compounds thus obtained from screening, are docked with both Human and Plasmodium DHODH. Potential anti-malarial lead compounds can be developed to treat resistant strains of Plasmodium falciparum.
{"title":"Development of specific DHODH inhibitors for Plasmodium and Human species","authors":"P. Swaminathan, L. Saleena","doi":"10.1504/IJCBDD.2019.098175","DOIUrl":"https://doi.org/10.1504/IJCBDD.2019.098175","url":null,"abstract":"Malaria still remains one of the challenging public health issue infecting about 300-500 millions of people. The most serious and fatal malarial infections are caused by Plasmodium falciparum which has developed resistance to commonly employed therapeutics. Hence the need to develop a novel anti-malarial drug targeting Dihydroorotate dehydrogenase (DHODH), an enzyme involved in parasite growth. DHODH is present in both humans and Plasmodium falciparum. Sequence analysis and structure comparison of DHODH of both Human and Plasmodium falciparum reveals variations among them, thereby providing a chance to design a specific inhibitor. Virtual screening of existing anti-malarial drugs acting on DHODH is performed from Pubchem and BindingDB databases. Pharmacophore mapping was done for the top 20 virtual screening compounds using hip hop algorithm. The compounds thus obtained from screening, are docked with both Human and Plasmodium DHODH. Potential anti-malarial lead compounds can be developed to treat resistant strains of Plasmodium falciparum.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"18 1","pages":"1-15"},"PeriodicalIF":0.0,"publicationDate":"2019-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74803911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1504/IJCBDD.2019.10021272
Yan-Shuo Chu, Mingxiang Teng, Yadong Wang
{"title":"Simulating genetically heterozygous genomes in the tumour tissue according to its clonal evolution history","authors":"Yan-Shuo Chu, Mingxiang Teng, Yadong Wang","doi":"10.1504/IJCBDD.2019.10021272","DOIUrl":"https://doi.org/10.1504/IJCBDD.2019.10021272","url":null,"abstract":"","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"150 1","pages":"143-152"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84977108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1504/ijcbdd.2019.10022513
Meriem Meyar, S. Feddal, Zohra Bouakouk, S. Kellou-Tairi
{"title":"Exploration of cyclooxygenase-1 binding modes of some chiral anti-inflammatory drugs using molecular docking and dynamic simulations","authors":"Meriem Meyar, S. Feddal, Zohra Bouakouk, S. Kellou-Tairi","doi":"10.1504/ijcbdd.2019.10022513","DOIUrl":"https://doi.org/10.1504/ijcbdd.2019.10022513","url":null,"abstract":"","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"10 1","pages":"281-301"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74292555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-14DOI: 10.1504/IJCBDD.2018.096126
I. Almasri
Natural products have long been considered as important sources for drug discovery due to the diversity of their chemical structures and broad range of biological activities attained by modulation of different biological targets. Therefore, the identification of the molecular targets of natural products is a milestone step in rational design of more potent and safer compounds. In this work, we explored the polypharmacology of three natural products having pleiotropic health beneficial effects: resveratrol, curcumin and berberine, using a ligand-based target fishing approach. The fishing protocol was started with the generation of a chemogenomic database that links individual targets with specific target ligands or group of drugs. Targets profile was then generated using ROCS software. The applied method was able not only to retrieve known targets within the top-ranked list for the natural compounds but also identified off-targets which were found by docking simulation to be potential targets and were consistent with recently identified bioactivities of these compounds.
{"title":"Exploring polypharmacology of some natural products using similarity search target fishing approach","authors":"I. Almasri","doi":"10.1504/IJCBDD.2018.096126","DOIUrl":"https://doi.org/10.1504/IJCBDD.2018.096126","url":null,"abstract":"Natural products have long been considered as important sources for drug discovery due to the diversity of their chemical structures and broad range of biological activities attained by modulation of different biological targets. Therefore, the identification of the molecular targets of natural products is a milestone step in rational design of more potent and safer compounds. In this work, we explored the polypharmacology of three natural products having pleiotropic health beneficial effects: resveratrol, curcumin and berberine, using a ligand-based target fishing approach. The fishing protocol was started with the generation of a chemogenomic database that links individual targets with specific target ligands or group of drugs. Targets profile was then generated using ROCS software. The applied method was able not only to retrieve known targets within the top-ranked list for the natural compounds but also identified off-targets which were found by docking simulation to be potential targets and were consistent with recently identified bioactivities of these compounds.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"28 1","pages":"295-309"},"PeriodicalIF":0.0,"publicationDate":"2018-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81713908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-14DOI: 10.1504/IJCBDD.2018.096125
Madhulata Kumari, Neeraj Tiwari, N. Subbarao
Feature selection approaches have been widely applied to deal with the various sample size problem in the classification of activity of datasets. The present work focuses on the understanding system of descriptors of anti-malarial inhibitors by Genetic programming (GP) to understand the impact of descriptors on inhibitory effects. The experimental dataset of inhibitors of anti-malarial was used to derive the optimised system by GP. Additionally, we have developed machine learning models using the random forest, decision tree, support vector machine (SVM) and Naive Bayes on an antimalarial dataset obtained from ChEMBL database and evaluated for their predictive capability. Based on the statistical evaluation, Random Forest model showed the higher area under the curve (AUC), better accuracy, sensitivity, and specificity in the cross-validation tests as compared to others. The statistical results indicated that the RF model was the best predictive model with 82.51% accuracy, 89.7% ROC. We deployed the RF classifier model on three datasets; phytochemical compound dataset, NCI natural product dataset IV and approved drugs dataset containing 918, 423 and 1554 compounds resulting 153, 81 and 250 compounds respectively as anti-malarial compounds. Further, to prioritise drug-like compounds, Lipinski's rule was applied on active phytochemicals which resulted in 13 hit anti-malarial molecules. Thus, such predictive models are useful to find out novel hit anti-malarial compounds and could also be used to discover novel drugs for other diseases.
{"title":"A genetic programming-based approach and machine learning approaches to the classification of multiclass anti-malarial datasets","authors":"Madhulata Kumari, Neeraj Tiwari, N. Subbarao","doi":"10.1504/IJCBDD.2018.096125","DOIUrl":"https://doi.org/10.1504/IJCBDD.2018.096125","url":null,"abstract":"Feature selection approaches have been widely applied to deal with the various sample size problem in the classification of activity of datasets. The present work focuses on the understanding system of descriptors of anti-malarial inhibitors by Genetic programming (GP) to understand the impact of descriptors on inhibitory effects. The experimental dataset of inhibitors of anti-malarial was used to derive the optimised system by GP. Additionally, we have developed machine learning models using the random forest, decision tree, support vector machine (SVM) and Naive Bayes on an antimalarial dataset obtained from ChEMBL database and evaluated for their predictive capability. Based on the statistical evaluation, Random Forest model showed the higher area under the curve (AUC), better accuracy, sensitivity, and specificity in the cross-validation tests as compared to others. The statistical results indicated that the RF model was the best predictive model with 82.51% accuracy, 89.7% ROC. We deployed the RF classifier model on three datasets; phytochemical compound dataset, NCI natural product dataset IV and approved drugs dataset containing 918, 423 and 1554 compounds resulting 153, 81 and 250 compounds respectively as anti-malarial compounds. Further, to prioritise drug-like compounds, Lipinski's rule was applied on active phytochemicals which resulted in 13 hit anti-malarial molecules. Thus, such predictive models are useful to find out novel hit anti-malarial compounds and could also be used to discover novel drugs for other diseases.","PeriodicalId":13612,"journal":{"name":"Int. J. Comput. Biol. Drug Des.","volume":"46 1","pages":"275-294"},"PeriodicalIF":0.0,"publicationDate":"2018-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88961147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：