In Silico Pharmacokinetic Profiling, ADME Prediction, and Drug-likeness Evaluation of Phenolic Constituents Identified in Honey: Implications for Asthma and Inflammatory Disorders

1. Introduction
Honey is one of the oldest natural products used for medicinal purposes, with therapeutic applications spanning several millennia across diverse civilizations [1, 2]. Beyond its nutritional value, honey has traditionally been employed in the management of wounds, gastrointestinal disorders, respiratory conditions, and infections [1]. Its antioxidant, antimicrobial, anti-inflammatory, and antiproliferative properties are primarily attributed to phytochemical constituents, particularly flavonoids and phenolic acids [2, 3]. The composition of these bioactive compounds varies considerably with botanical and geographical origin, thereby influencing therapeutic potential [4, 5]. Honey phenolics comprise two major groups: phenolic acids and flavonoids. Phenolic acids include hydroxybenzoic acid derivatives, such as gallic acid, p-hydroxybenzoic acid, and vanillic acid, and hydroxycinnamic acid derivatives, including caffeic, p-coumaric, ferulic, and cinnamic acids. Flavonoids include flavones (chrysin, apigenin), flavonols (quercetin, kaempferol, galangin), flavanones (pinocembrin, naringenin, hesperetin), and flavanonols (pinobanksin). Their relative abundance depends on nectar sources and is influenced by geographical, climatic, and seasonal factors [3-6]. These compounds can modulate oxidative stress, inflammatory signaling, and immune responses through multiple mechanisms [2, 3].
Despite the growing literature on honey-derived phenolics, pharmacokinetic and drug-likeness evaluations have primarily focused on compounds identified in European, Mediterranean, Asian, and Australasian honey varieties [7]. Recent analyses of honey samples from southern Kazakhstan identified fumaric acid, p-hydroxybenzoic acid, p-coumaric acid, trans-2-hydroxycinnamic acid, and chrysin as commonly occurring phenolics across multiple sampling localities [7]. Plant-derived polyphenols have attracted increasing attention because of their pleiotropic pharmacological activities, favorable preclinical safety profiles, and long history of human consumption [1-5]. Their ability to interact with multiple inflammatory pathways may be advantageous in complex multifactorial diseases. Phenolic acids and flavonoids have been reported to suppress Th2 cytokine production, inhibit mast cell degranulation, reduce eosinophil recruitment, and attenuate mucus hypersecretion in experimental models of allergic airway inflammation and asthma [6, 7].
Honey has traditionally been used for the symptomatic management of respiratory conditions, including cough, bronchitis, and asthma [7, 8]. Its demulcent properties may alleviate bronchial irritation, while its phenolic constituents may modulate airway inflammation through antioxidant and anti-inflammatory mechanisms [7, 8]. Several honey-derived phenolics suppress NF-kB and MAPK signaling, reduce pro-inflammatory cytokine production, and attenuate oxidative stress in preclinical models [9-11]. For example, chrysin attenuates IgE-mediated immune responses and airway hyperresponsiveness, whereas p-coumaric acid inhibits cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) expression through NF-kB pathway suppression [9-11].
In silico approaches have become indispensable in contemporary drug discovery by enabling rapid assessment of absorption, distribution, metabolism, excretion properties before experimental validation [12-15]. Computational platforms such as SwissADME facilitate prediction of physicochemical characteristics, drug-likeness, and pharmacokinetic behavior, supporting the identification of compounds with favorable pharmaceutical profiles [14, 15]. Therefore, the present study aimed to evaluate the drug-likeness potential and pharmacokinetic profiles of five phenolic compounds commonly identified in honey in the context of inflammatory disorders.

2. Materials and Methods
2.1. Honey samples and compound selection
Honey samples were collected from four distinct localities in southern Kazakhstan (Table 1).

Based on analytical findings and literature data, five biologically active compounds consistently detected across the samples were selected for in silico evaluation: fumaric acid, p-hydroxybenzoic acid, p-coumaric acid, trans-2-hydroxycinnamic acid, and chrysin. The phenolic profile of Kazakhstani honey varies with nectar source, geographical origin, and climatic conditions. These compounds were selected because they were repeatedly detected in honey samples from different locations in southern Kazakhstan [4, 5]. Here, “commonly identified” refers to their recurrence across independent samples rather than relative abundance. Other constituents were excluded because they were detected only in specific samples, occurred at trace levels, or lacked sufficient consistency for comparative in silico analysis. The selected compounds also represent major phenolic classes in honey, including hydroxybenzoic acids, hydroxycinnamic acids, and flavonoids, providing a structurally diverse and representative subset for preliminary pharmacokinetic and drug-likeness screening.

2.2. In silico pharmacokinetic and drug-likeness assessment
The Simplified Molecular Input Line Entry System (SMILES) notations for each compound were retrieved from the PubChem database (National Center for Biotechnology Information, U.S. National Library of Medicine) [15, 16] and are presented in Table 2.

The physicochemical, pharmacokinetic, and drug-likeness properties of the selected compounds were evaluated using the SwissADME web platform, a freely accessible computational tool developed by the Swiss Institute of Bioinformatics [15, 16]. Toxicity parameters, including LD₅₀ values and toxicity classes, were determined using the ProTox-3.0 online server [15, 16]. Bioavailability radar plots were generated to provide a graphical overview of oral bioavailability-related properties across six dimensions: lipophilicity (LIPO), molecular size (SIZE), polarity (POLAR), solubility (INSOLU), saturation (INSATU), and flexibility (FLEX) [15-19]. The BOILED-Egg (Brain or IntestinaL EstimateD permeation) model was used to visualize predicted passive GI absorption and BBB permeation based on WLOGP and TPSA coordinates [16, 17].

2.3. Molecular docking studies
The potential binding modes and intermolecular interactions of the bioactive compounds with the target protein were investigated using Maestro 14.5 (Schrödinger, 2025–3). The three-dimensional crystal structure was retrieved from the Protein Data Bank (PDB) under accession code 5KIR. Prior to docking, the protein was prepared using the Protein Preparation Wizard in Schrödinger Maestro 14.5, including addition of missing hydrogen atoms, assignment of partial charges, completion of missing side chains, and modeling of absent loop regions. Water molecules located more than 3 Å from the binding site were removed. The prepared structure was energy-minimized at pH 7 using the OPLS4e force field. Ligands were prepared with LigPrep by optimizing ionization states within pH 7±2, verifying stereochemistry, and performing energy minimization using the OPLS3e force field. The receptor grid was generated around the protein active site, with standardized dimensions of 20×20×20 Å. For 5KIR, grid-center coordinates were defined according to the active site, and docking simulations were performed around these coordinates [18-20].

3. Results
3.1. Physicochemical properties
The physicochemical properties of the five bioactive compounds are summarized in Table 3.

Molecular weights ranged from 116.07 g/mol (fumaric acid) to 254.24 g/mol (chrysin), with all compounds below the Lipinski threshold of 500 g/mol [21]. Chrysin exhibited the highest number of aromatic heavy atoms (16 of 19), greatest molar refractivity (71.97), and most extensive ring system, indicating substantial potential for van der Waals and π–π stacking interactions with biological targets. TPSA values ranged from 57.53 Ų (p-hydroxybenzoic acid, p-coumaric acid, and trans-2-hydroxycinnamic acid) to 74.60 Ų (fumaric acid), remaining below the 140 Ų threshold associated with poor intestinal absorption [22]. All compounds had two hydrogen-bond donors and 3–4 acceptors, satisfying Lipinski criteria and indicating favorable intermolecular interaction potential [21]. Fsp3 was 0.00 for all compounds, reflecting their predominantly aromatic or unsaturated character and falling below the recommended 0.25 saturation threshold of the SwissADME bioavailability radar [14].
P-coumaric acid and trans-2-hydroxycinnamic acid shared the molecular formula C9H8O3 and molecular weight (164.16 g/mol) but differed in hydroxyl position. Although this isomerism does not alter the global physicochemical descriptors calculated by SwissADME, it may affect intramolecular hydrogen bonding, conformation, and biological activity [14, 21-23]. Overall, chrysin and the hydroxycinnamic acid derivatives showed the most favorable profiles for biological target engagement.
These differences reflect the structural characteristics of the compounds. Fumaric acid, a non-aromatic dicarboxylic acid, has high polarity and hydrogen-bonding capacity but limited hydrophobic and π-stacking interactions. P-hydroxybenzoic acid contains one aromatic ring with carboxylic acid and phenolic hydroxyl groups, resulting in moderate aromaticity and lipophilicity. The hydroxycinnamic acids, p-coumaric acid and trans-2-hydroxycinnamic acid, contain a propenoic acid side chain that increases conjugation and conformational rigidity. Chrysin is the most structurally complex compound, comprising three fused rings and two phenolic hydroxyl groups at C-5 and C-7 that may facilitate specific hydrogen-bonding interactions with molecular targets. These structural differences are reflected in their physicochemical profiles in molecular weight, molar refractivity (Table 3).

3.2. Lipophilicity analysis
Lipophilicity is a key determinant of membrane permeability and oral bioavailability [22, 23]. Consensus Log P values, calculated from five computational models, are presented in Table 4 and Figure 1. 

.Chrysin showed the highest value (2.55), consistent with its extended aromatic system and relatively low polar surface area, suggesting favorable passive membrane permeation and hydrophobic protein interactions [21-23]. Trans-2-Hydroxycinnamic acid (1.40) and p-coumaric acid (1.26) fell within the optimal range for oral drug candidates (1–3), indicating balanced hydrophilic-lipophilic properties [23-25]. P-hydroxybenzoic acid showed moderate lipophilicity (1.05), whereas fumaric acid had a negative log P (-0.35), reflecting its highly hydrophilic character. Although all compounds fell within the broadly acceptable range (-0.7 to +5), fumaric acid may exhibit limited passive diffusion across lipid bilayers [15].

3.3. Aqueous solubility
The predicted aqueous solubility values obtained using ESOL, Ali, and SILICOS-IT are summarized in Table 4. Fumaric acid showed the highest predicted solubility, classified as “very soluble” by ESOL (log S=-0.21) and Ali (log S=-0.76), consistent with its low molecular weight and high polarity. p-Hydroxybenzoic acid, p-coumaric acid, and trans-2-hydroxycinnamic acid were classified as “soluble” by all three models, indicating favorable dissolution properties for oral administration [22-24].
Chrysin exhibited the lowest predicted solubility and was classified as “moderately soluble” (ESOL log S=-4.19; Ali log S=-4.69; SILICOS-IT log S=-4.98), likely reflecting its extended aromatic conjugation and higher lipophilicity. Thus, despite its favorable membrane-permeation and protein-binding characteristics, poor aqueous solubility may limit its oral bioavailability and require formulation strategies such as nanoencapsulation, cyclodextrin complexation, or solid dispersion [22-24]. The hydroxycinnamic acid derivatives showed a favorable balance between solubility and lipophilicity, suggesting potentially more predictable pharmacokinetic behavior (Table 4).

3.4. Pharmacokinetic predictions
The predicted pharmacokinetic properties are presented in Table 5.

All five compounds showed high predicted gastrointestinal (GI) absorption, consistent with their favorable Lipinski profiles and TPSA values [22-25]. Four compounds were predicted to cross the blood-brain barrier (BBB), whereas fumaric acid was not, consistent with its higher polarity (TPSA=74.60 Ų) and low lipophilicity in the BOILED-Egg model. None of the compounds were predicted to be p-glycoprotein (P-gp) substrates or inhibitors of CYP1A2, CYP2C19, CYP2C9, CYP2D6, or CYP3A4. These predictions suggest a low likelihood of active efflux and CYP-mediated pharmacokinetic interactions, although experimental validation is required [25]. Chrysin showed the highest predicted skin permeation coefficient (log Kp=-5.35 cm/s), whereas fumaric acid showed the lowest (-7.25 cm/s), consistent with their lipophilicity profiles (Figures 2 and 3) [25].

3.5. Drug-likeness assessment
Drug-likeness was evaluated using five complementary filters (Lipinski, Ghose, Veber, Egan, and Muegge), with results presented in Table 6.

All five compounds satisfied the Lipinski, Veber, and Egan criteria without violations, supporting their basic suitability for oral drug development [22-25]. P-coumaric acid and trans-2-hydroxycinnamic acid also passed the Ghose filter, whereas fumaric acid and p-hydroxybenzoic acid showed Ghose and Muegge violations, mainly due to their low molecular weights [23-25]. Chrysin was the only compound satisfying all five drug-likeness filters. However, its predicted bioavailability score (0.55) was lower than that of the other compounds (0.85), likely reflecting its lower aqueous solubility (Figure 3). This finding highlights the importance of the solubility–lipophilicity balance for oral bioavailability and the potential need for formulation optimization of chrysin [25].

3.6. Medicinal chemistry attributes
Medicinal chemistry assessment (Table 6) showed that none of the five compounds triggered PAINS alerts, suggesting a low likelihood of nonspecific reactivity [25]. Brenk michael_acceptor_1 alerts were identified for fumaric acid, p-coumaric acid, and trans-2-hydroxycinnamic acid due to their α,β-unsaturated carbonyl groups. Although this feature may indicate electrophilic reactivity, it is also common in approved drugs and natural products and does not preclude further development [25].
Synthetic accessibility scores ranged from 1 (p-hydroxybenzoic acid) to 2.93 (chrysin) on a 1–10 scale, indicating relatively favorable synthetic accessibility for all compounds (Figure 3). Combined with their natural occurrence in honey, these findings support their potential for further development [25].

​​​​​​​3.7. Integrated BOILED-Egg and bioavailability radar interpretation
The BOILED-Egg model predicts gastrointestinal absorption and BBB permeation based on WLOGP and TPSA coordinates [23-25]. As shown in Figure 2, all five compounds were located in the white region, indicating high predicted human intestinal absorption, while four were also positioned in the yellow region (yolk), suggesting potential passive BBB permeation. Fumaric acid was outside the yolk, consistent with its high polarity and TPSA. All compounds were PGP−, indicating no predicted p-glycoprotein substrate activity.
Although BBB permeability may be relevant to neuroimmune interactions, its significance here should be interpreted cautiously because these findings are based solely on computational predictions. Accordingly, BBB permeability should be considered a general pharmacokinetic indicator rather than evidence of therapeutic central effects.
The bioavailability radar plots showed that p-coumaric acid and trans-2-hydroxycinnamic acid had the most balanced profiles across the evaluated parameters. All compounds failed the saturation criterion, reflecting their predominantly aromatic structures. Overall, integrated physicochemical, pharmacokinetic, and drug-likeness analyses identified p-coumaric acid and trans-2-hydroxycinnamic acid as the most favorable candidates, followed by p-hydroxybenzoic acid, chrysin, and fumaric acid (Figure 3).

3.8. Molecular prediction and molecular docking results
Molecular docking revealed marked differences in binding affinities among the compounds (Table 7).

Chrysin showed the strongest predicted affinity (−8.412 kcal/mol), likely due to its extensive flavonoid scaffold and favorable hydrophobic complementarity. Among the phenolic acids, p-hydroxybenzoic acid (−6.522 kcal/mol) showed stronger binding than trans-2-hydroxycinnamic acid (−5.794 kcal/mol) and p-coumaric acid (−5.633 kcal/mol), whereas fumaric acid exhibited the weakest affinity (−3.331 kcal/mol), consistent with its smaller size and limited interaction capacity.
Chrysin formed hydrophobic interactions with Val523, Ala527, and Leu352, while p-hydroxybenzoic acid established polar contacts with Arg120 and Tyr355. Trans-2-hydroxycinnamic acid and p-coumaric acid interacted with Arg120 through their acrylate side chains, with p-coumaric acid additionally interacting with Ser353. Fumaric acid interacted with Ser530 and Gly526 within a confined region of the active site (Figure 4).

Overall, the docking results indicate that complementary hydrophobic and hydrogen-bonding interactions contribute to ligand stabilization within the target binding site.

4. Discussion
The pharmacokinetic and drug-likeness profiles established in the present study gain particular significance when considered alongside the reported biological activities of these compounds. Fumaric acid, through derivatives such as dimethyl fumarate (DMF), activates the nuclear factor erythroid 2-related factor 2 (Nrf2) pathway, enhancing antioxidant defenses while suppressing NF-kB-mediated pro-inflammatory cytokine production [6, 7]. This dual mechanism is highly relevant to oxidative stress-related diseases, including inflammation and asthma. P-coumaric acid suppresses COX-2 and iNOS expression via inhibition of NF-kB and MAPK signaling and reduces tumor necrosis factor (TNF)-alpha (TNF-α) and interleukin 6 (IL-6) production in inflammatory models [6-8]. Its balanced lipophilicity and high predicted GI absorption support the feasibility of these effects following oral administration. Similarly, trans-2-hydroxycinnamic acid modulates inflammatory pathways and exhibits antioxidant activity, with a pharmacokinetic profile comparable to that of p-coumaric acid [9-12]. Chrysin warrants particular attention due to its ability to suppress NF-kB, STAT3, and PI3K/Akt signaling, modulate mast cell activation, and attenuate IgE-mediated immune responses [6-12]. Experimental asthma models have shown reductions in airway hyperresponsiveness and eosinophilic infiltration [6-12]. Although its high molar refractivity and lipophilicity suggest strong receptor-binding potential, limited aqueous solubility remains a key pharmacokinetic challenge that may be overcome through nanocarrier systems or structural modifications. p-Hydroxybenzoic acid exhibits free radical scavenging activity through its phenolic hydroxyl group. Although related hydroxybenzoic acid derivatives reduce pro-inflammatory mediator expression [5-12], direct evidence for p-hydroxybenzoic acid remains limited. The coexistence of these five phenolics in honey suggests potential synergistic interactions. Multi-target modulation of the NF-kB, MAPK, and PI3K/Akt pathways may provide broader anti-inflammatory effects than single-target approaches [5-12]. Their diverse physicochemical properties, ranging from hydrophilic fumaric acid to lipophilic chrysin, indicate complementary biodistribution patterns that may contribute to the enhanced biological activity often observed in whole honey compared with isolated constituents. Overall, the ADME and drug-likeness analyses indicate that p-coumaric acid, trans-2-hydroxycinnamic acid, and chrysin possess favorable pharmacokinetic characteristics for modulating inflammatory signaling. Their multi-target mechanisms, natural origin, and reported preclinical safety support further evaluation as complementary therapeutic candidates for inflammatory diseases. Several limitations should be acknowledged. SwissADME predictions do not fully capture the complexity of in vivo pharmacokinetics, including first-pass metabolism, microbiome-mediated biotransformation, and tissue-specific accumulation. Likewise, the BOILED-Egg model predicts BBB permeability based solely on passive diffusion and excludes several active transport mechanisms. In addition, commonly used drug-likeness filters were developed primarily using synthetic compounds and may have limited applicability to low-molecular-weight natural phenolics. Future studies should validate these findings through experimental ADME approaches, including Caco-2 permeability, microsomal stability, and plasma protein binding assays.

5. Conclusion
The present in silico investigation demonstrated that the five phenolic compounds commonly identified in honey samples from the southern regions of Kazakhstan possess favorable physicochemical, pharmacokinetic, and drug-likeness profiles that suggest their potential as candidates for further experimental investigations. It should be emphasized that the present findings are based exclusively on computational predictions and require experimental confirmation. Future studies incorporating molecular docking simulations, in vitro cell culture models, and in vivo animal models are warranted to validate the hypotheses generated in this study. 

Acknowledgements
The authors would like to express their gratitude to Khoja Akhmet Yassawi International Kazakh-Turkish University, Niğde Ömer Halisdemir University, and Sivas Cumhuriyet University for institutional support. The authors also thank the beekeepers of the Turkistan region for their cooperation in the collection of propolis samples and for sharing their practical expertise.

Compliance with ethical guidelines
We hereby declare all ethical standards have been respected in preparation of the submitted article. 

Funding
This study was supported by the project on the development of innovative technologies and creation of scientific infrastructure for the sustainable development of the South Kazakhstan region, funded by the Scientific Committee of the Ministry of Science and Higher Education of the Republic of Kazakhstan, Astana, Kazakhstan (Code: BR24992814).

Authors' contributions
Conceptualization and study design: Sevgi Durna Dastan and Zeliha Selamoglu; Data collection: Eda Sonmez Gurer and Taner Dastan; Formal analysis and interpretation: Eda Sonmez Gurer and Sevgi Durna Dastan. Writing the original draft: Sevgi Durna Dastan, Nurdana Salybekova, and Zeliha Selamoglu; Review and editing: Zeliha Selamoglu and Ainash Oshibayeva; Administrative, technical, and material support: Nurdana Salybekova, Ainash Oshibayeva, Sevgi Durna Dastan, and Zeliha Selamoglu.

Conflict of interest
The authors declared no conflict of interest.

Data availability
All data generated or analyzed during this study are included in this published article. The SwissADME platform is freely accessible. Compound SMILES notations were retrieved from the PubChem database.

References

1. Samarghandian S, Farkhondeh T, Samini F. Honey and health: A review of recent clinical research. Pharmacognosy Res. 2017; 9(2):121-7. [DOI:10.4103/0974-8490.204647] [PMID]
2. Cianciosi D, Forbes-Hernandez TY, Afrin S, Gasparrini M, Reboredo-Rodriguez P, Manna PP, et al. Phenolic compounds in honey and their associated health benefits: A review. Molecules. 2018; 23(9):2322. [DOI:10.3390/molecules23092322] [PMID]
3. Alvarez-Suarez JM, Gasparrini M, Forbes-Hernandez TY, Mazzoni L, Giampieri F. The composition and biological activity of honey: A focus on Manuka honey. Foods. 2014; 3(3):420-32. [DOI:10.3390/foods3030420] [PMID]
4. Kedzierska-Matysek M, Stryjecka M, Teter A, Skalecki P, Domaradzki P, Florek M. Relationships between the content of phenolic compounds and the antioxidant activity of Polish honey varieties as tool botanical discrimination. Molecules. 2021; 26(6):1810. [DOI:10.3390/molecules26061810] [PMID]
5. Ongalbek D, Tokul-Olmez O, Sahin B, Kucukaydin S, Aydogmus-Ozturk F, Sicak Y, et al. Classification of buckwheat honey produced in Kazakhstan according to their biochemical ingredients and bioactivities by chemometric approach. Food Chem. 2024; 451:139409. [DOI:10.1016/j.foodchem.2024.139409] [PMID]
6. Papi A, Brightling C, Pedersen SE, Reddel HK. Asthma. Lancet. 2018; 391(10122):783-800. [DOI:10.1016/S0140-6736(17)33311-1] [PMID]
7. Barnes PJ. Targeting cytokines to treat asthma and chronic obstructive pulmonary disease. Nat Rev Immunol. 2018; 18(7):454-66. [DOI:10.1038/s41577-018-0006-6] [PMID]
8. Pei K, Ou J, Huang J, Ou S. p-Coumaric acid and its conjugates: Dietary sources, pharmacokinetic properties biological activities. J Sci Food Agric. 2016; 96(9):2952-62. [DOI:10.1002/jsfa.7578] [PMID]
9. Neog MK, Joshua Pragasam S, Krishnan M, Rasool M. p-Coumaric acid, a dietary polyphenol ameliorates inflammation and curtails cartilage and bone erosion in the rheumatoid arthritis rat model. Biofactors. 2017; 43(5):698-717. [PMID]
10. Taofiq O, González-Paramás AM, Barreiro MF, Ferreira ICFR. Hydroxycinnamic acids and their derivatives: Cosmeceutical significance, challenges and future perspectives, a review. Molecules. 2017; 22(2):281. [DOI:10.3390/molecules22020281] [PMID]
11. Yao J, Jiang M, Zhang Y, Liu X, Du Q, Feng G. Chrysin alleviates allergic inflammation and airway remodeling in a murine model of chronic asthma. Int Immunopharmacol. 2016; 32:24-31. [DOI:10.1016/j.intimp.2016.01.005] [PMID]
12. Sliwoski G, Kothiwale S, Meiler J, Lowe EW Jr. Computational methods in drug discovery. Pharmacol Rev. 2014; 66(1):334-95. [DOI:10.1124/pr.112.007336] [PMID]
13. Ferreira LLG, Andricopulo AD. ADMET modeling approaches in drug discovery. Drug Discov Today. 2019; 24(5):1157-65. [DOI:10.1016/j.drudis.2019.03.015] [PMID]
14. Daina A, Michielin O, Zoete V. SwissADME: A free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep. 2017; 7:42717. [DOI:10.1038/srep42717] [PMID]
15. Banerjee P, Eckert AO, Schrey AK, Preissner R. ProTox-II: A webserver for the prediction of toxicity of chemicals. Nucleic Acids Res. 2018; 46(W1):W257-63. [DOI:10.1093/nar/gky318] [PMID]
16. Kim S, Chen J, Cheng T, Gindulyte A, He J, He S, et al. PubChem 2023 update. Nucleic Acids Res. 2023; 51(D1):D1373-80. [DOI:10.1093/nar/gkac956] [PMID]
17. Daina A, Zoete V. A BOILED-Egg to predict gastrointestinal absorption and brain penetration of small molecules. ChemMedChem. 2016; 11(11):1117-21. [DOI:10.1002/cmdc.201600182] [PMID]
18. Schrödinger: LigPrep. New York: Schrödinger; 2025. [Link]
19. Schrödinger: Maestro. New York: Schrödinger; 2025. [Link]
20. Schrödinger. Protein preparation workflow; Epik; Impact; Prime. New York: Schrödinger; 2025. [Link]
21. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev. 2001; 46(1-3):3-26. [DOI:10.1016/S0169-409X(00)00129-0] [PMID]
22. Veber DF, Johnson SR, Cheng HY, Smith BR, Ward KW, Kopple KD. Molecular properties that influence the oral bioavailability of drug candidates. J Med Chem. 2002; 45(12):2615-23. [DOI:10.1021/jm020017n] [PMID]
23. Arnott JA, Planey SL. The influence of lipophilicity in drug discovery and design. Expert Opin Drug Discov. 2012; 7(10):863-75. [DOI:10.1517/17460441.2012.714363] [PMID]
24. Delaney JS. ESOL: Estimating aqueous solubility directly from molecular structure. J Chem Inf Comput Sci. 2004; 44(3):1000-5. [DOI:10.1021/ci034243x] [PMID]
25. Mani R, Natesan V. Chrysin: Sources, beneficial pharmacological activities, and molecular mechanism action. Phytochemistry. 2018; 145:187-96. [DOI:10.1016/j.phytochem.2017.09.016] [PMID]