DDDevendra Kumar Dhaked

Devendra Kumar Dhaked

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Devendra Kumar Dhaked has a total of 43 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Interpretable Machine Learning for Predicting Cytochrome P450 Mechanism-Based Inhibition

    by Khushashish Bhasin, Sudham Engle, Devendra Kumar Dhaked

    Cytochrome P450 (CYP450) enzymes play a crucial role in the metabolism of endogenous and exogenous compounds, and their inhibition can lead to clinically significant drug–drug interactions and adverse effects. Mechanism-based inhibition (MBI) is a time-dependent form of CYP450 inhibition in which enzyme-mediated bioactivation of an inhibitor can result in persistent or irreversible enzyme inactivation. Early identification of compounds with MBI liability is therefore important for reducing metabolic and drug–drug interaction risks during drug discovery. In this study, a curated dataset of 569 unique MBI compounds was compiled from the literature, while non-mechanism-based inhibitors (NMBIs) were obtained from PubChem. To construct a chemically diverse negative dataset, Butina clustering-based sampling was employed. Molecular representations based on RDKit and Mordred descriptors, MACCS keys, and ECFP fingerprints were evaluated using five machine learning algorithms, with feature selection and model optimization applied systematically. CYP-aware stratified splitting was employed to ensure appropriate representation of CYP isoform–class combinations across the training and test sets, and its performance was compared with random stratified splitting. Among the evaluated molecular representations, RDKit descriptors generally provided the strongest predictive performance, with optimized models achieving test-set ROC-AUC values of up to 0.92 in the CYP-aware evaluation. Applicability-domain analysis further demonstrated higher prediction accuracy for compounds within the defined chemical space, supporting the reliability of predictions for structurally represented compounds. In addition, structural-alert analysis identified class-associated structural motifs, including statistically significant alerts preferentially associated with MBI compounds, providing chemically interpretable information complementary to the machine learning models. Evaluation using the expanded MBI569 dataset further assessed the robustness of the predictive models across different dataset sizes and splitting strategies. Overall, this study demonstrates a systematic and interpretable approach for predicting CYP450 mechanism-based inhibition, with potential utility for the early identification of MBI liabilities during drug discovery.

  • Open Access

    Modulating the reactivity of anionic phosphido‑boranes via π‑conjugated spacers for reversible CO2 capture

    by Tejas Bhute, Arya Kc, Devendra Kumar Dhaked

    Carbon dioxide (CO 2 ) capture using phosphine–borane-based systems has emerged as a promising strategy for sustainable carbon utilization and activation. In this study, density functional theory calculations were employed to investigate alkali-metal-assisted azo- and acetylene-bridged phosphine–borane systems for CO 2 capture and activation. The metal-free systems exhibited weak or thermodynamically unfavorable CO 2 complexation, whereas incorporation of alkali metals significantly enhanced CO 2 stabilization, with Li + consistently outperforming the corresponding Na + and K + analogues. Electronic structure and NBO analyses revealed that the acetylene-linked systems possess greater phosphorus nucleophilicity and more efficient electronic communication than their azo-linked counterparts. Reaction-profile analyses showed that CO 2 activation proceeds through favorable geometric distortion and strong donor–acceptor interactions involving the phosphorus center, activated CO 2 fragment, and alkali-metal cation. Substituent effects revealed that modification at the boron centre effectively tunes CO 2 binding, whereas substitution at phosphorus generally weakens or suppresses CO 2 capture by reducing phosphorus nucleophilicity. Among the systems investigated, acetylene-linked and Li+-assisted derivatives exhibited slightly more favorable CO 2 binding characteristics. Furthermore, the calculated reaction profiles indicate that CO 2 release is associated with moderate energy barriers, suggesting the potential for reversible CO 2 capture and efficient regeneration. These findings highlight the critical roles of linker design, alkali-metal coordination, and substituent engineering in developing efficient phosphine–borane systems for CO 2 activation.

  • UV Light-Mediated Generation of Ketyl Radicals with a Concurrent 1,2-Boron Shift and One-Pot Synthesis of Benzoxaboroles under Continuous-Flow Conditions

    by Kankanala Naveen Kumar, Chandrakant Sahu, Pranjal Mahindrakar, Devendra Kumar Dhaked, Sharada Prasanna Swain

    The photocatalyzed reactions using metal and organic photocatalysts generate ketyl radicals by a single-electron transfer, proton-coupled electron transfer mechanism. Photocatalyst-free UV light-mediated generation of ketyl radicals in the presence of bis(pinacoloto)diboron (B2pin2) is developed. The addition of the Bpin radical to a carbonyl group followed by a 1,2-boron shift generated a thermodynamically favorable ketyl radical. This method was applied to the one-pot synthesis of benzoxaboroles from o-halo benzaldehydes.


  • One-pot cascade synthesis of 2-aminomaleimides and thioaminomaleimides: an alternative fluorogenic platform capable of drug conjugation

    by Harendra Kumar, Aakriti Garg, Sawan Kumar, Devendra Kumar Dhaked, Sreya Gupta, Subhendu Bhowmik

    A DPPA-driven cascade enables the formation of fluorogenic aminomaleimides, thioaminomaleimides, and drug-conjugated amidomaleimides from maleamic acids.


  • Open Access

    State of the art of Machine learning for Targeted ProteinDegradation: Molecular Glues Prediction

    by Dikshit Rathva, Sudham Engle, Devendra Kumar Dhaked

    Targeted protein degradation is a rapidly evolving field in drug discovery, particularly for diseases with undruggable targets. Molecular glues (MGS) are a new class of small molecules that induce two proteins to come together, leading to the degradation of the protein of interest by an E3 ligase. The identification of such molecules has traditionally relied on serendipitous discovery or labour-intensive experimental screening. Their simple structure and broad applicability make them a cornerstone for the next generation of drug discovery. To overcome these limitations, this study integrates machine learning (ML)-based prediction for the in silico identification of potential molecular glue (MG). A dataset of 2091 MG and 196 non-molecular glue (NMG) molecules was curated from literature, patent sources, and databases. A wide range of 2D (and 3D) molecular descriptors from RDKit, Mordred, PaDEL, Dragon, alvaDesc and PyDisciptors were used for the development of machine learning models. Three feature selection methods (RFE, RFECV, Boruta) and five algorithms (RF, XGBoost, SVM, Naïve Bayes, k-NN) were employed. Results from these models showed that machine learning can be used for predicting MG molecules, with 80% of models exhibiting an ROC-AUC greater than 0.90 and 78% of models achieving an F1 score greater than 0.90. XGBoost and RF consistently demonstrated high performance across all descriptors, with an ROC-AUC of over 0.95. However, the deep learning-based D-MPNN model slightly outperformed both, indicating its enhanced capacity for capturing complex molecular patterns. An analysis of physicochemical parameters revealed that MGs tend to be more lipophilic and structurally heavier, with an increased polar surface area and a higher number of hydrogen bond acceptors, compared to typical FDA-approved drugs. Multiple structural alerts, which are specific to MGs, were identified with a p-value of <0.05. Out of 54 identified SAs, the top 10 have very high accuracy and a low positive rate; of these, two SAs were present in over 400 MGs.

  • Open Access

    Mechanistic Insights into the Antioxidant Activity of Baicalein and Related Flavones: A DFT and Molecular Dynamics Study

    by Nidhi Hismatray Aswani, Sawan Kumar, Devendra Kumar Dhaked

    This study presents a comprehensive density functional theory (DFT) investigation of the antioxidant mechanisms of baicalein and four related flavones: chrysin, wogonin, norwogonin, and oroxylin A. Using the M06-2X/6-311+G(d,p) level of theory, we evaluated their radical scavenging potentials through multiple thermodynamic descriptors associated with different antioxidant pathways, including hydrogen atom transfer (HAT), sequential proton loss–electron transfer (SPLET), single electron transfer–proton transfer (SETPT), sequential proton loss–hydrogen atom transfer (SPLHAT), and double hydrogen atom transfer (dHAT). The conformational analysis revealed that intramolecular hydrogen bonding significantly stabilizes geometries, influencing reactivity. Among all investigated compounds, baicalein and norwogonin, each containing three phenolic groups, exhibited the most favourable bond dissociation enthalpies (BDEs), proton affinities (PAs) and ionisation potentials (IPs), particularly at the 6-OH and 8-OH positions, respectively. The study identifies key structure–activity relationships, emphasizing the role of hydroxyl positions and their count as well as phenyl ring in modulating antioxidant efficiency. Molecular dynamics-derived solubility profiles reveal the preferential compatibility of flavones with nonpolar solvents, emphasizing the critical influence of the phenyl ring in modulating their solubility behavior. These findings establish a theoretical foundation for the design of potent antioxidant compounds.


  • Exploring natural phenolic compounds as amyloid-beta fibril inhibitors: computational insights for Alzheimer’s disease

    by Rakibul Islam, Devendra Kumar Dhaked

    Alzheimer's disease (AD) is a progressive neurodegenerative disorder and remains one of the leading causes of death in older age. While there is no direct cure for AD, very few FDA-approved drugs are available for managing the symptoms by targeting traditional targets. Amyloid-beta (Aß) peptides were identified as a crucial target for AD therapy, in recent times, phenolic compounds have shown the potential to inhibit Amyloid-beta peptide. These phenolic compounds destabilize Aß aggregation, but their inhibition mechanisms are not clearly understood. In this study, we thoroughly explored the molecular binding mechanism of >1000 phenolic compounds with the Aß fibrils via extra precision (XP) and induced fit docking (IFD) methods. Additionally, molecular dynamics (MD) simulations of the top 10 druglike phenolic compounds were performed with a simulation time of 500 nanoseconds (ns). Our results showed that two promising compounds, Rhapontigenin and Okanin, exhibited superior binding free energy and stability compared to the reference apigenin. These compounds effectively disrupted the Asp23-Lys28 salt bridge interactions and hydrogen bond patterns in the Aß fibril. Furthermore, both the identified compounds showed significant van der Waal's interaction with the Aß fibril. Our findings align with the previous studies and provide additional insights into inhibitory mechanisms at the molecular level. Additionally, this study incorporates analysis of non-covalent interactions and DFT studies to further elucidate the mechanisms. Nevertheless, it is crucial to emphasize that these results are based on in silico analyses and necessitate further experimental validation through in vitro and in vivo studies to confirm their efficacy.


  • Tautomeric Conflicts in Forty Small-Molecule Databases

    by Devendra K. Dhaked, Marc C. Nicklaus

    We have analyzed 40 different databases ranging in size from a few thousand to nearly 100 million molecules, comprising a total of over 210 million structures, for their tautomeric conflicts. A tautomeric conflict is defined as an occurrence of two or more structures within a data set identified by the tautomeric rules applied as being tautomers of each other. We tested a total of 119 detailed tautomeric transform rules expressed as SMIRKS, out of which 79 yielded at least one conflict. These transformations include three types of tautomerism: prototropic, ring-chain, and valence tautomerism. The databases analyzed spanned a wide variety of types including large aggregating databases, drug collections, and structure collections based on experimental data. All databases analyzed showed intra-database tautomeric conflicts. The conflict rates as percentage of the database were typically in the few tenths of a percent range, which for the largest databases amounts to >100,000 cases per database.


  • Open Access

    What impact does tautomerism have on drug properties and development?

    by Devendra Dhaked, Marc Nicklaus

    This mini-review reports on effects that tautomerism has on predicted and clinically observed small-molecule properties important for drug design and drug behaviour in the patient. We present examples of both pharmacokinetics and pharmacodynamics effects. The importance of tautomerism on drugs’ formulation and possible interconversion of stereoisomers is also discussed.