Predicting the multiple parameters of organic acceptors through machine learning using RDkit descriptors: An easy and fast pipeline DOI Open Access
Khadijah Mohammedsaleh Katubi, Muhammad Saqib,

Tayyaba Mubashir

et al.

International Journal of Quantum Chemistry, Journal Year: 2023, Volume and Issue: 123(23)

Published: Aug. 25, 2023

Abstract Machine learning (ML) analysis has gained huge importance among researchers for predicting multiple parameters and designing efficient donor acceptor materials without experimentation. Data are collected from literature subsequently used impactful properties of organic solar cells such as power conversion efficiency (PCE) energy levels (HOMO/LUMO). Importantly, out various tested models, hist gradient boosting (HGB) the light (LGBM) regression models revealed better predictive capabilities. To achieve prediction effectively, selected (best) ML further tuned. For PCE (test set), LGBM shows coefficient determination ( R 2 ) value 0.787, which is higher than HGB = 0.680). HOMO 0.566, 0.563). However, LUMO 0.605, lower 0.606). Among three predicted properties, ability PCE. These help to predict acceptors in a short time less computational cost.

Language: Английский

Easy and fast prediction of green solvents for small molecule donor-based organic solar cells through machine learning DOI
Asif Mahmood, Yahya Sandali, Jin‐Liang Wang

et al.

Physical Chemistry Chemical Physics, Journal Year: 2023, Volume and Issue: 25(15), P. 10417 - 10426

Published: Jan. 1, 2023

Solubility plays a critical role in many aspects of research (drugs to materials). parameters are very useful for selecting appropriate solvents/non-solvents various applications. In the present study, Hansen solubility predicted using machine learning. More than 40 models tried search best model. Molecular descriptors and fingerprints used as inputs get comparative view. Machine learning trained molecular have shown higher prediction ability model fingerprints. their potential be easy fast compared density functional theory (DFT)/thermodynamic approach. creates "black box" connection properties. Therefore, minimal computational cost is required. With help best-trained model, green solvents selected small molecule donors that organic solar cells. Our introduced framework can select cells an way.

Language: Английский

Citations

102
Machine learning assisted designing of Y-series small molecule acceptors: Library generation and property prediction DOI
Farooq Ahmad, Asif Mahmood, Islam H. El Azab

et al.

Journal of Photochemistry and Photobiology A Chemistry, Journal Year: 2024, Volume and Issue: 453, P. 115670 - 115670

Published: April 12, 2024

Language: Английский

Citations

29
Virtual screening and library enumeration of new hydroxycinnamates based antioxidant compounds: A complete framework DOI Creative Commons
Jameel Ahmed Bhutto,

Tayyaba Mubashir,

Mudassir Hussain Tahir

et al.

Journal of Saudi Chemical Society, Journal Year: 2023, Volume and Issue: 27(4), P. 101670 - 101670

Published: June 7, 2023

Designing of molecules for drugs is important topic from many decades. The search new very hard, and it expensive process. Computer assisted framework can provide the fastest way to design screen drug-like compounds. In present work, a multidimensional approach introduced designing screening antioxidant Antioxidants play crucial role in ensuring that body's oxidizing reducing species are kept proper balance, minimizing oxidative stress. Machine learning models used predict activity. Three hydroxycinnamates selected as standard antioxidants. Similar compounds searched ChEMBL database using chemical structural similarity method. libraries generated evolutionary New also designed automatic decomposition construction building blocks. activity all predicted machine models. space envisioned t-distributed stochastic neighbor embedding (t-SNE) Best shortlisted, their synthetic accessibility further facilitate experimental chemists. between studied fingerprints heatmap.

Language: Английский

Citations

30
Unveiling potent inhibitors for schistosomiasis through ligand-based drug design, molecular docking, molecular dynamics simulations and pharmacokinetics predictions DOI Creative Commons
Saudatu Chinade Ja’afaru,

Adamu Uzairu,

İmren Bayıl

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(6), P. e0302390 - e0302390

Published: June 26, 2024

Schistosomiasis is a neglected tropical disease which imposes considerable and enduring impact on affected regions, leading to persistent morbidity, hindering child development, diminishing productivity, imposing economic burdens. Due the emergence of drug resistance limited management options, there need develop additional effective inhibitors for schistosomiasis. In view this, quantitative structure-activity relationship studies, molecular docking, dynamics simulations, drug-likeness pharmacokinetics predictions were applied 39 Schistosoma mansoni Thioredoxin Glutathione Reductase (SmTGR) inhibitors. The chosen QSAR model demonstrated robust statistical parameters, including an R 2 0.798, adj 0.767, Q cv 0.681, LOF 0.930, test 0.776, cR p 0.746, confirming its reliability. most active derivative (compound 40 ) was identified as lead candidate development new potential non-covalent through ligand-based design. Subsequently, 12 novel compounds ( 40a-40l designed with enhanced anti-schistosomiasis activity binding affinity. Molecular docking studies revealed strong stable interactions, hydrogen bonding, between target receptor. simulations over 100 nanoseconds MM-PBSA free energy (ΔG bind calculations validated stability two best-designed molecules. Furthermore, prediction analyses affirmed these compounds, suggesting their promise innovative agents treatment

Language: Английский

Citations

13
Optimizing the performance of phase-change azobenzene: from trial and error to machine learning DOI
Kai Wang, Huitao Yu,

Jing-Li Gao

et al.

Journal of Materials Chemistry C, Journal Year: 2024, Volume and Issue: 12(11), P. 3811 - 3837

Published: Jan. 1, 2024

Machine learning can predict the properties of phase change azobenzene derivatives and guide molecular design to further improve their photothermal conversion performance.

Language: Английский

Citations

9
TCMBank: bridges between the largest herbal medicines, chemical ingredients, target proteins, and associated diseases with intelligence text mining DOI Creative Commons
Qiujie Lv, Guanxing Chen, Haohuai He

et al.

Chemical Science, Journal Year: 2023, Volume and Issue: 14(39), P. 10684 - 10701

Published: Jan. 1, 2023

Traditional Chinese Medicine (TCM) has long been viewed as a precious source of modern drug discovery. AI-assisted discovery (AIDD) investigated extensively. However, there are still two challenges in applying AIDD to guide TCM discovery: the lack large amount standardized TCM-related information and is prone pathological failures out-of-domain data. We have released Database@Taiwan 2011, it widely disseminated used. Now, we developed TCMBank, largest systematic free database, which an extension Database@Taiwan. TCMBank contains 9192 herbs, 61 966 ingredients (unduplicated), 15 179 targets, 32 529 diseases, their pairwise relationships. By integrating multiple data sources, provides 3D structure standard list detailed on ingredients, targets diseases. intelligent document identification module that continuously adds retrieved from literature PubChem. In addition, driven by big data, ensemble learning-based protocol for identifying potential leads repurposing. take colorectal cancer Alzheimer's disease examples demonstrate how accelerate artificial intelligence. Using researchers can view literature-driven relationship mapping between herbs/ingredients genes/diseases, allowing understanding molecular action mechanisms new potentially effective treatments. available at https://TCMBank.CN/.

Language: Английский

Citations

20
Designing of symmetric and asymmetric small molecule acceptors for organic solar cells: A farmwork based on Machine learning, virtual screening and structural analysis DOI

Tayyaba Mubashir,

Mudassir Hussain Tahir, M.H.H. Mahmoud

et al.

Journal of Photochemistry and Photobiology A Chemistry, Journal Year: 2023, Volume and Issue: 444, P. 114977 - 114977

Published: June 24, 2023

Language: Английский

Citations

19
Data mining and library generation to search electron-rich and electron-deficient building blocks for the designing of polymers for photoacoustic imaging DOI Creative Commons
Muhammad Ishfaq,

Tayyaba Mubashir,

Safaa N. Abdou

et al.

Heliyon, Journal Year: 2023, Volume and Issue: 9(11), P. e21332 - e21332

Published: Oct. 26, 2023

Photoacoustic imaging is a good method for biological imaging, this purpose, materials with strong near infrared (NIR) absorbance are required. In the present study, machine learning models used to predict light absorption behavior of polymers. Molecular descriptors utilized train variety models. Building blocks searched from chemical databases, as well new building designed using library enumeration method. The Breaking Retrosynthetically Interesting Chemical Substructures (BRICS) employed creation 10,000 novel These polymers based on input and selected blocks. To enhance process, optimal model UV/visible maxima newly Concurrently, similarity analysis also performed polymers, synthetic accessibility calculated. summary, all easy synthesize, increasing their potential practical applications.

Language: Английский

Citations

18
Synthesis of 1,2,3-triazole-piperazin-benzo[b][1,4]thiazine 1,1-dioxides: antibacterial, hemolytic and in silico TLR4 protein inhibitory activities DOI Creative Commons

N. Ramu,

Thupurani Murali Krishna, Ravikumar Kapavarapu

et al.

RSC Advances, Journal Year: 2024, Volume and Issue: 14(13), P. 8921 - 8931

Published: Jan. 1, 2024

Novel 1,2,3-triazoles (6a–6j & 8a–8g) were synthesized and evaluated for their antibacterial activity against S. aureus . The more potent compounds further in silico TLR4 inhibitory activity.

Language: Английский

Citations

7
Computational design of new polymers having low exciton binding energy for organic solar cells fabrication: Chemical generation and visualization DOI
Fatimah Mohammed Alzahrani, Alvi Muhammad Rouf, Jawayria Najeeb

et al.

Journal of Photochemistry and Photobiology A Chemistry, Journal Year: 2024, Volume and Issue: 450, P. 115457 - 115457

Published: Jan. 4, 2024

Language: Английский

Citations

5