Journal of Computer-Aided Molecular Design最新文献_第8页

On the construction of LIECE models for the serotonin receptor 5-HT(_{text {2A}})R 5-羟色胺受体5-HT LIECE模型的构建(_{text {2A}}) R

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-06-14 DOI: 10.1007/s10822-023-00507-3

Aida Shahraki, Jana Selent, Peter Kolb

引用次数: 0

Integrated data-driven and experimental approaches to accelerate lead optimization targeting SARS-CoV-2 main protease 整合数据驱动和实验方法，加速针对SARS-CoV-2主要蛋白酶的导联优化

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-06-14 DOI: 10.1007/s10822-023-00509-1

Rohith Anand Varikoti, Katherine J. Schultz, Chathuri J. Kombala, Agustin Kruel, Kristoffer R. Brandvold, Mowei Zhou, Neeraj Kumar

{"title":"Integrated data-driven and experimental approaches to accelerate lead optimization targeting SARS-CoV-2 main protease","authors":"Rohith Anand Varikoti, Katherine J. Schultz, Chathuri J. Kombala, Agustin Kruel, Kristoffer R. Brandvold, Mowei Zhou, Neeraj Kumar","doi":"10.1007/s10822-023-00509-1","DOIUrl":"10.1007/s10822-023-00509-1","url":null,"abstract":"<div><p>Identification of potential therapeutic candidates can be expedited by integrating computational modeling with domain aware machine learning (ML) models followed by experimental validation in an iterative manner. Generative deep learning models can generate thousands of new candidates, however, their physiochemical and biochemical properties are typically not fully optimized. Using our recently developed deep learning models and a scaffold as a starting point, we generated tens of thousands of compounds for SARS-CoV-2 M<sup>pro</sup> that preserve the core scaffold. We utilized and implemented several computational tools such as structural alert and toxicity analysis, high throughput virtual screening, ML-based 3D quantitative structure-activity relationships, multi-parameter optimization, and graph neural networks on generated candidates to predict biological activity and binding affinity in advance. As a result of these combined computational endeavors, eight promising candidates were singled out and put through experimental testing using Native Mass Spectrometry and FRET-based functional assays. Two of the tested compounds with quinazoline-2-thiol and acetylpiperidine core moieties showed IC<span>(_{50})</span> values in the low micromolar range: <span>(2.95pm 0.0017)</span> <span>(upmu)</span>M and 3.41±0.0015 <span>(upmu)</span>M, respectively. Molecular dynamics simulations further highlight that binding of these compounds results in allosteric modulations within the chain B and the interface domains of the M<sup>pro</sup>. Our integrated approach provides a platform for data driven lead optimization with rapid characterization and experimental validation in a closed loop that could be applied to other potential protein targets.</p></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 8","pages":"339 - 355"},"PeriodicalIF":3.5,"publicationDate":"2023-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4576444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

OFraMP: a fragment-based tool to facilitate the parametrization of large molecules OFraMP:一个基于片段的工具，便于大分子的参数化

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-06-13 DOI: 10.1007/s10822-023-00511-7

Martin Stroet, Bertrand Caron, Martin S. Engler, Jimi van der Woning, Aude Kauffmann, Marc van Dijk, Mohammed El-Kebir, Koen M. Visscher, Josef Holownia, Callum Macfarlane, Brian J. Bennion, Svetlana Gelpi-Dominguez, Felice C. Lightstone, Tijs van der Storm, Daan P. Geerke, Alan E. Mark, Gunnar W. Klau

{"title":"OFraMP: a fragment-based tool to facilitate the parametrization of large molecules","authors":"Martin Stroet, Bertrand Caron, Martin S. Engler, Jimi van der Woning, Aude Kauffmann, Marc van Dijk, Mohammed El-Kebir, Koen M. Visscher, Josef Holownia, Callum Macfarlane, Brian J. Bennion, Svetlana Gelpi-Dominguez, Felice C. Lightstone, Tijs van der Storm, Daan P. Geerke, Alan E. Mark, Gunnar W. Klau","doi":"10.1007/s10822-023-00511-7","DOIUrl":"10.1007/s10822-023-00511-7","url":null,"abstract":"<div><p>An Online tool for Fragment-based Molecule Parametrization (OFraMP) is described. OFraMP is a web application for assigning atomic interaction parameters to large molecules by matching sub-fragments within the target molecule to equivalent sub-fragments within the Automated Topology Builder (ATB, atb.uq.edu.au) database. OFraMP identifies and compares alternative molecular fragments from the ATB database, which contains over 890,000 pre-parameterized molecules, using a novel hierarchical matching procedure. Atoms are considered within the context of an extended local environment (buffer region) with the degree of similarity between an atom in the target molecule and that in the proposed match controlled by varying the size of the buffer region. Adjacent matching atoms are combined into progressively larger matched sub-structures. The user then selects the most appropriate match. OFraMP also allows users to manually alter interaction parameters and automates the submission of missing substructures to the ATB in order to generate parameters for atoms in environments not represented in the existing database. The utility of OFraMP is illustrated using the anti-cancer agent paclitaxel and a dendrimer used in organic semiconductor devices.</p><h3>Graphical abstract</h3><p>OFraMP applied to paclitaxel (ATB ID 35922).</p><figure><div><div><div><picture><source><img></source></picture></div></div></div></figure></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 8","pages":"357 - 371"},"PeriodicalIF":3.5,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10822-023-00511-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4541924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Molecular dynamics simulations reveal the inhibition mechanism of Cdc42 by RhoGDI1 分子动力学模拟揭示了RhoGDI1对Cdc42的抑制机制

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-06-07 DOI: 10.1007/s10822-023-00508-2

Yijing Zhang, Shiyao Chen, Taeyoung Choi, Yuzheng Qi, Qianhui Wang, Guanyi Li, Yaxue Zhao

{"title":"Molecular dynamics simulations reveal the inhibition mechanism of Cdc42 by RhoGDI1","authors":"Yijing Zhang, Shiyao Chen, Taeyoung Choi, Yuzheng Qi, Qianhui Wang, Guanyi Li, Yaxue Zhao","doi":"10.1007/s10822-023-00508-2","DOIUrl":"10.1007/s10822-023-00508-2","url":null,"abstract":"<div><p>Cell division control protein 42 homolog (Cdc42), which controls a variety of cellular functions including rearrangements of the cell cytoskeleton, cell differentiation and proliferation, is a potential cancer therapeutic target. As an endogenous negative regulator of Cdc42, the Rho GDP dissociation inhibitor 1 (RhoGDI1) can prevent the GDP/GTP exchange of Cdc42 to maintain Cdc42 into an inactive state. To investigate the inhibition mechanism of Cdc42 through RhoGDI1 at the atomic level, we performed molecular dynamics (MD) simulations. Without RhoGDI1, Cdc42 has more flexible conformations, especially in switch regions which are vital for binding GDP/GTP and regulators. In the presence of RhoGDI1, it not only can change the intramolecular interactions of Cdc42 but also can maintain the switch regions into a closed conformation through extensive interactions with Cdc42. These results which are consistent with findings of biochemical and mutational studies provide deep structural insights into the inhibition mechanisms of Cdc42 by RhoGDI1. These findings are beneficial for the development of novel therapies targeting Cdc42-related cancers.</p></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 7","pages":"301 - 312"},"PeriodicalIF":3.5,"publicationDate":"2023-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4305400","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Elucidating the potential effects of point mutations on FGFR3 inhibitor resistance via combined molecular dynamics simulation and community network analysis 通过结合分子动力学模拟和社区网络分析阐明点突变对FGFR3抑制剂耐药性的潜在影响

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-06-03 DOI: 10.1007/s10822-023-00510-8

Bo Liu, Juntao Ding, Yugang Liu, Jianzhang Wu, Xiaoping Wu, Qian Chen, Wulan Li

{"title":"Elucidating the potential effects of point mutations on FGFR3 inhibitor resistance via combined molecular dynamics simulation and community network analysis","authors":"Bo Liu, Juntao Ding, Yugang Liu, Jianzhang Wu, Xiaoping Wu, Qian Chen, Wulan Li","doi":"10.1007/s10822-023-00510-8","DOIUrl":"10.1007/s10822-023-00510-8","url":null,"abstract":"<div><p>FGFR3 kinase mutations are associated with a variety of malignancies, but FGFR3 mutant inhibitors have rarely been studied. Furthermore, the mechanism of pan-FGFR inhibitors resistance caused by kinase domain mutations is still unclear. In this study, we try to explain the mechanism of drug resistance to FGFR3 mutation through global analysis and local analysis based on molecular dynamics simulation, binding free energy analysis, umbrella sampling and community network analysis. The results showed that FGFR3 mutations caused a decrease in the affinity between drugs and FGFR3 kinase, which was consistent with the reported experimental results. Possible mechanisms are that mutations affect drug-protein affinity by altering the environment of residues near the hinge region where the protein binds to the drug, or by affecting the A-loop and interfering with the allosteric communication networks. In conclusion, we systematically elucidated the underlying mechanism of pan-FGFR inhibitor resistance caused by FGFR3 mutation based on molecular dynamics simulation strategy, which provided theoretical guidance for the development of FGFR3 mutant kinase inhibitors.</p></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 7","pages":"325 - 338"},"PeriodicalIF":3.5,"publicationDate":"2023-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10822-023-00510-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4128652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Insights into the coordination chemistry of antineoplastic doxorubicin with 3d-transition metal ions Zn2+, Cu2+, and VO2+: a study using well-calibrated thermodynamic cycles and chemical interaction quantum chemistry models 抗肿瘤药物多柔比星与 3d 过渡金属离子 Zn2+、Cu2+ 和 VO2+ 的配位化学透视：使用校准良好的热力学循环和化学相互作用量子化学模型进行的研究

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-05-28 DOI: 10.1007/s10822-023-00506-4

Julieta Reyna-Luna, Luis Soriano-Agueda, Christiaan Jardinez Vera, Marco Franco-Pérez

{"title":"Insights into the coordination chemistry of antineoplastic doxorubicin with 3d-transition metal ions Zn2+, Cu2+, and VO2+: a study using well-calibrated thermodynamic cycles and chemical interaction quantum chemistry models","authors":"Julieta Reyna-Luna, Luis Soriano-Agueda, Christiaan Jardinez Vera, Marco Franco-Pérez","doi":"10.1007/s10822-023-00506-4","DOIUrl":"10.1007/s10822-023-00506-4","url":null,"abstract":"<div><p>We present a computational strategy based on thermodynamic cycles to predict and describe the chemical equilibrium between the 3<i>d</i>-transition metal ions Zn<sup>2+</sup>, Cu<sup>2+</sup>, and VO<sup>2+</sup> and the widely used antineoplastic drug doxorubicin. Our method involves benchmarking a theoretical protocol to compute gas-phase quantities using DLPNO Coupled-Cluster calculations as reference, followed by estimating solvation contributions to the reaction Gibbs free energies using both explicit partial (micro)solvation steps for charged solutes and neutral coordination complexes, as well as a continuum solvation procedure for all solutes involved in the complexation process. We rationalized the stability of these doxorubicin-metal complexes by inspecting quantities obtained from the topology of their electron densities, particularly the bond critical points and non-covalent interaction index. Our approach allowed us to identify representative species in solution phase, infer the most likely complexation process for each case, and identify key intramolecular interactions involved in the stability of these compounds. To the best of our knowledge, this is the first study reporting thermodynamic constants for the complexation of doxorubicin with transition metal ions. Unlike other methods, our procedure is computationally affordable for medium-sized systems and provides valuable insights even with limited experimental data. Furthermore, it can be extended to describe the complexation process between 3<i>d-</i>transition metal ions and other bioactive ligands.\u0000</p></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 7","pages":"279 - 299"},"PeriodicalIF":3.5,"publicationDate":"2023-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"5091618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Study of SQ109 analogs binding to mycobacterium MmpL3 transporter using MD simulations and alchemical relative binding free energy calculations 利用MD模拟和炼金术相对结合自由能计算研究SQ109类似物与分枝杆菌MmpL3转运体的结合

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-05-02 DOI: 10.1007/s10822-023-00504-6

Marianna Stampolaki, Ioannis Stylianakis, Helen I. Zgurskaya, Antonios Kolocouris

{"title":"Study of SQ109 analogs binding to mycobacterium MmpL3 transporter using MD simulations and alchemical relative binding free energy calculations","authors":"Marianna Stampolaki, Ioannis Stylianakis, Helen I. Zgurskaya, Antonios Kolocouris","doi":"10.1007/s10822-023-00504-6","DOIUrl":"10.1007/s10822-023-00504-6","url":null,"abstract":"<div><p><i>N</i>-geranyl-<i>N</i>΄-(2-adamantyl)ethane-1,2-diamine (SQ109) is a tuberculosis drug that has high potency against <i>Mycobacterium tuberculosis (Mtb)</i> and may function by blocking cell wall biosynthesis. After the crystal structure of MmpL3 from <i>Mycobacterium smegmatis</i> in complex with SQ109 became available, it was suggested that SQ109 inhibits Mmpl3 mycolic acid transporter. Here, we showed using molecular dynamics (MD) simulations that the binding profile of nine SQ109 analogs with inhibitory potency against Mtb and alkyl or aryl adducts at C-2 or C-1 adamantyl carbon to MmpL3 was consistent with the X-ray structure of MmpL3 – SQ109 complex. We showed that rotation of SQ109 around carbon–carbon bond in the monoprotonated ethylenediamine unit favors two <i>gauche</i> conformations as minima in water and lipophilic solvent using DFT calculations as well as inside the transporter’s binding area using MD simulations. The binding assays in micelles suggested that the binding affinity of the SQ109 analogs was increased for the larger, more hydrophobic adducts, which was consistent with our results from MD simulations of the SQ109 analogues suggesting that sizeable C-2 adamantyl adducts of SQ109 can fill a lipophilic region between Y257, Y646, F260 and F649 in MmpL3. This was confirmed quantitatively by our calculations of the relative binding free energies using the thermodynamic integration coupled with MD simulations method with a mean assigned error of 0.74 kcal mol<sup>−1</sup> compared to the experimental values.</p><h3>Graphical abstract</h3>\u0000 <figure><div><div><div><picture><source><img></source></picture></div></div></div></figure>\u0000 </div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 5-6","pages":"245 - 264"},"PeriodicalIF":3.5,"publicationDate":"2023-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10822-023-00504-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4097786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

TargIDe: a machine-learning workflow for target identification of molecules with antibiofilm activity against Pseudomonas aeruginosa TargIDe:一种机器学习工作流程，用于对铜绿假单胞菌具有抗生素膜活性的分子进行目标识别

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-04-22 DOI: 10.1007/s10822-023-00505-5

João Carneiro, Rita P. Magalhães, Victor M. de la Oliva Roque, Manuel Simões, Diogo Pratas, Sérgio F. Sousa

{"title":"TargIDe: a machine-learning workflow for target identification of molecules with antibiofilm activity against Pseudomonas aeruginosa","authors":"João Carneiro, Rita P. Magalhães, Victor M. de la Oliva Roque, Manuel Simões, Diogo Pratas, Sérgio F. Sousa","doi":"10.1007/s10822-023-00505-5","DOIUrl":"10.1007/s10822-023-00505-5","url":null,"abstract":"<div><p>Bacterial biofilms are a source of infectious human diseases and are heavily linked to antibiotic resistance. <i>Pseudomonas aeruginosa</i> is a multidrug-resistant bacterium widely present and implicated in several hospital-acquired infections. Over the last years, the development of new drugs able to inhibit <i>Pseudomonas aeruginosa</i> by interfering with its ability to form biofilms has become a promising strategy in drug discovery. Identifying molecules able to interfere with biofilm formation is difficult, but further developing these molecules by rationally improving their activity is particularly challenging, as it requires knowledge of the specific protein target that is inhibited. This work describes the development of a machine learning multitechnique consensus workflow to predict the protein targets of molecules with confirmed inhibitory activity against biofilm formation by <i>Pseudomonas aeruginosa</i>. It uses a specialized database containing all the known targets implicated in biofilm formation by <i>Pseudomonas aeruginosa.</i> The experimentally confirmed inhibitors available on ChEMBL, together with chemical descriptors, were used as the input features for a combination of nine different classification models, yielding a consensus method to predict the most likely target of a ligand. The implemented algorithm is freely available at https://github.com/BioSIM-Research-Group/TargIDe under licence GNU General Public Licence (GPL) version 3 and can easily be improved as more data become available.</p></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 5-6","pages":"265 - 278"},"PeriodicalIF":3.5,"publicationDate":"2023-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10822-023-00505-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4845270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Computational insights into ligand–induced G protein and β-arrestin signaling of the dopamine D1 receptor 配体诱导的G蛋白和多巴胺D1受体的β-阻滞蛋白信号的计算见解

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-04-15 DOI: 10.1007/s10822-023-00503-7

Haoxi Li, Nikhil M. Urs, Nicole Horenstein

{"title":"Computational insights into ligand–induced G protein and β-arrestin signaling of the dopamine D1 receptor","authors":"Haoxi Li, Nikhil M. Urs, Nicole Horenstein","doi":"10.1007/s10822-023-00503-7","DOIUrl":"10.1007/s10822-023-00503-7","url":null,"abstract":"<div><p>The dopamine D1 receptor (D1R), is a class A G protein coupled-receptor (GPCR) which has been a promising drug target for psychiatric and neurological disorders such as Parkinson’s disease (PD). Previous studies have suggested that therapeutic effects can be realized by targeting the β-arrestin signaling pathway of dopamine receptors, while overactivation of the G protein-dependent pathways leads to side effects, such as dyskinesias. Therefore, it is highly desirable to develop a D1R ligand that selectively regulates the β-arrestin pathway. Currently, most D1R agonists are signaling-balanced and stimulate both G protein and β-arrestin pathways, with a few reports of G protein biased ligands. However, identification and characterization of β-arrestin biased D1R agonists has been a challenge thus far. In this study, we implemented Gaussian accelerated molecular dynamics (GaMD) simulations to provide valuable computational insights into the possible underlying molecular mechanism of the different signaling properties of two catechol and two non-catechol D1R agonists that are either G protein biased or signaling-balanced. Dynamic network analysis further identified critical residues in the allosteric signaling network of D1R for each ligand at different conformational or binding states. Some of these residues are crucial for G protein or arrestin signals of GPCRs based on previous studies. Finally, we provided a molecular design strategy which can be utilized by medicinal chemists to develop potential β-arrestin biased D1R ligands. The proposed hypotheses are experimentally testable and can guide the development of safer and more effective medications for a variety of CNS disorders.</p></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 5-6","pages":"227 - 244"},"PeriodicalIF":3.5,"publicationDate":"2023-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4597295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Improvement of multi-task learning by data enrichment: application for drug discovery 通过数据充实改进多任务学习:在药物发现中的应用

IF 3.5 3区生物学

Journal of Computer-Aided Molecular Design Pub Date : 2023-03-21 DOI: 10.1007/s10822-023-00500-w

Ekaterina A. Sosnina, Sergey Sosnin, Maxim V. Fedorov

{"title":"Improvement of multi-task learning by data enrichment: application for drug discovery","authors":"Ekaterina A. Sosnina, Sergey Sosnin, Maxim V. Fedorov","doi":"10.1007/s10822-023-00500-w","DOIUrl":"10.1007/s10822-023-00500-w","url":null,"abstract":"<div><p>Multi-task learning in deep neural networks has become a topic of growing importance in many research fields, including drug discovery. However, applying multi-task learning poses new challenges in improving prediction performance. This study investigated the potential of training data enrichment to enhance multi-task model prediction quality in drug discovery. The study evaluated four scenarios with varying degrees of information capacity of the training data and applied two types of test data to evaluate prediction performance. We used three datasets: ViralChEMBL, which consisted of binary activities of compounds against viral species, was applied for the classification task; pQSAR(159) and pQSAR(4267), which consisted of bio-activities of compounds and assays from the research of the profile-QSAR method, were applied for regression tasks. We built multi-task models based on the feed-forward DNNs using the PyTorch framework. Our findings showed that training data enrichment could be an effective means of enhancing prediction performance in multi-task learning, but the degree of improvement depends on the quality of the training data. The more unique compounds and targets the training data included, the more new compound-target interactions are required for prediction improvement. Also, we found out that even using multi-task learning, one could not predict the interactions of compounds that are highly dissimilar from those used for model training. The study provides some recommendations for effectively employing multi-task learning in drug discovery to improve prediction accuracy and facilitate the discovery of novel drug candidates.</p></div>","PeriodicalId":621,"journal":{"name":"Journal of Computer-Aided Molecular Design","volume":"37 4","pages":"183 - 200"},"PeriodicalIF":3.5,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"4830796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4