Computational Identification of Small-Molecule Inhibitors Targeting TEAD2 Palmitoylation Pocket in Lung Cancer

INTRODUCTION

Lung cancer is among the most common and life-threatening form of cancer worldwide, with over one million individuals diagnosed each year. It’s a significant lethal burden related to the diseases. In spite of many significant advancements in cancer diagnosis and therapeutic interventions, lung cancer remains one of commonly primary causes of cancer-associated deaths globally (Chaitanya et al., 2021). The disease occurs as a consequence of expanding uncontrolled proliferation of abnormal cells within the lung that have developed by capacity to invade adjacent tissues and to metastasize to other sites including the brain liver bones and adrenal glands. They are divided into Non-Small Cell Lung Cancer (NSCLC) and Small Cell Lung Cancer (SCLC), with NSCLC comprising around 85% of all diagnosed cancer cases (Smolarz et al., 2025). The molecular heterogeneity and complexity of lung cancer have been of the greatest barriers to effective treatment. Standard treatment options like chemotherapy, surgery, and radiotherapy are still extensively employed mainly with the disease stage and histological subtype of tumor. Surgery is currently believed to the best form of treatment yet when a localized tumor has been diagnosed at an early stage. Unfortunately, most of the patients present with advanced or metastatic disease, where surgical options are limited. Multidrug resistance and non-selective toxicity limit the efficacy of chemotherapy drugs like platinum chemotherapy. Although, treatment-associated toxicities and tumor recurrence continue to be significant limits, radiotherapy particularly plays a crucial role in management of disease, especially in patients with advanced stages (Debela et al., 2021; Zafar et al., 2025). The main factor of high death rate in lung cancer is mainly due to tumor heterogeneity and resistance of the tumor to current treatments. The evident limitations of current treatment strategies underline the need to discover other molecular targets that are involved in the development of lung cancer and the development of resistance to the therapies that are used. The relevance of dysregulated signaling networks in tumor genesis, proliferation, survival migration and metastasis have place an increased focus in molecular oncology (Choudhary et al., 2023; He et al., 2021).

The Hippo signaling pathway has been identified as a crucial regulator of cellular proliferation, apoptosis, tissue homeostasis, stem cell maintenance and organ size regulation among the various molecular signaling pathways linked to the development of lung cancer. By limiting excessive cellular growth and encouraging programmed cell death, the hippo pathway acts as a tumor suppressor mechanism under normal physiological conditions. Dysfunctioning of this pathway treated as common algorithm that has been implicated in tumorigenesis due to the promotion of cell proliferation and resistance to apoptosis caused by the abnormal activation of certain transcriptional regulators (Fu et al., 2022; Yu et al., 2015; Zhong et al., 2024). The Transcriptional co-activators Yes-Associated Protein (YAP) and transcriptional co-activator with PDZ-binding motif (TAZ), which interacts with members of The Transcriptional Enhanced Associate Domain (TEAD) family of transcription factors, are considered as the main downstream effectors of the hippo pathways. Increasing evidence supports the importance of dysregulating the YAP/TAZ-TEAD signaling pathway in the process of tumorigenesis of diverse malignant disease like lung cancer, breast cancer, colorectal cancer, pancreatic tumor and hepatocellular carcinoma (Lo et al., 2018; Yu et al., 2025; Zhao et al., 2020).

Transcriptional Enhanced Associate Domain

The TEAD family contains four highly conserved transcription factors, TEAD1, TEAD2, TEAD3 and TEAD4. This can activate the transcription of growth- and development related genes such as proliferation differentiation, survival and migration and Epithelial-Mesenchymal Transition (EMT). Recently, TEAD2 has become more promising therapeutic target owing to its key role in the oncogenic transcriptional regulating pathway mainly in tumor with hyperactive hippo signaling pathway (Cunningham et al., 2022; Joo et al., 2020; Wang et al., 2024; Zhou et al., 2016). But, due to the particularity of transcription factor that contains few well-defined ligand-binding domains and other prominent active sites attract the traditional design of small molecules inhibitors. The TEAD family was often studied as undruggable target (Tao et al., 2023). Moreover, the current studies have recognized a conserved hydrophobic palmitoylation pocket with the TEAD proteins that particularly played important role for maintaining stability of protein and transcriptional activity. This has welcomed new opportunities for intervention of therapeutic targets. TEAD activity and its interaction with carcinogenic partners are regulated by palmitoylation, a post-translational alteration involving the covalent attachment of palmitic acid (Gridnev et al., 2022; Hu et al., 2022). Therefore, inhibiting the TEAD palmitoylation pocket has emerged as a novel therapeutic approach that can stop the transcriptional activation and slows down the growth of tumor cells and thereby suppressing the progression of the ailment.

The crystal structure of TEAD2 containing a ligand-bound palmitoylation pocket (PDB ID: 6E5G) was selected. This pocket represents a promising druggable site because small molecules capable of binding within this hydrophobic cavity may destabilize TEAD2 function, inhibit YAP/TAZ interaction, and ultimately suppress downstream oncogenic transcriptional activity, providing a valuable structural basis for structure-based drug discovery. Therefore, the present study aims to identify potential small-molecule inhibitors targeting the TEAD2 palmitoylation pocket involved in lung cancer progression using a structure-based computational approach. The study employs virtual screening of compounds from the MCULE database, molecular docking against the TEAD2 protein structure (PDB ID: 6E5G), and comprehensive evaluation of selected compounds through ADMET profiling, toxicity and side-effect prediction, drug-likeness assessment, and cancer cell-line sensitivity analysis to identify promising therapeutic candidates for lung cancer treatment.

MATERIALS AND METHODS

Macromolecule Retrieval and Preparation

The three-dimensional structure of the target protein (PDB-6E5G) was retrieved from the Protein Data Bank (https://www.rcsb.org/) in PDB format. The downloaded macromolecule structure was prepared with the BIOVIA Discovery Studio Visualizer (version 2021) (Berman, 2000). The preparation steps involved the removal of unwanted molecules, including water and heteroatoms, followed by hydrogen bond optimization and structural refinement to obtain an error-free structure for docking studies.

Structure-Based Virtual Screening of Small Molecules

Structure-based virtual screening was employed using the online Mcule platform considered as a comprehensive platform for computer aided drug discovery that combines multiple approaches including target identification, ligand-based screening, one-click docking, and access to commercially available compounds (Kiss et al., 2012). The hit-identification workflow was chosen within the structure-based virtual screening module of Mcule to screen out the potential molecules. Lipinski rule of five violation criteria was set between zero to one ensuring robust drug-likeness (Muteeb et al., 2022). The sampler option from the diversity section was opted. Rest remaining parameters were kept under pre-set values. For conducting the docking studies, the pre-prepared structure was uploaded and studies were conducted using the Autodock Vina enabled screening workflow. For grid box values, coordinates of the protein-ligand binding region were set (Trott et al., 2010; Roskoski et al., 2023).

Docking Evaluation of Inhibitor

Additionally, to validate the docking efficiency, the reference inhibitor compound was subjected to docked using one-click docking module under the category of lead optimization of Mcule platform. for that, processed macromolecule and inhibitor compound were uploaded and the binding dimensions were defined as per their complex coordinates. Top 4 poses were generated. The pose exhibiting the best energy and binding site was selected for further evaluations.

Structural Toxicity Screening

The shortlisted compounds were additionally evaluated using the Toxicity checker integrated within the Mcule platform (Kiss et al., 2012). Each molecule was screened by selecting the Mcule compound IDs. The result was categorized as pass or fail, indicating toxicity and safer compounds.

Study of Cancerous Cell Line Activity

The anticancer potential of filtered compounds was analyzed by the online CLC-Pred server, an online tool used for predicting the cytotoxicity effects against various human cell lines based on structure similarity. SMILES notations were submitted individually for each compound to generate the output depicted in the probability of activity and inactivity. Furthermore, inferred antitumor activity values were also generated, indicating greater therapeutic potential (Bishwakarma et al., 2026; Cheng et al., 2024; Lagunin et al., 2023).

Biological Activity Spectra Profiles

A biological activity spectrum was performed using the online server PASS (prediction of activity spectra for substances), which can predict 300+ pharmacological functions based on molecular using SMILES notation (SP, 2011). The SMILES were obtained from the Mcule platform and submitted to the server. The predicted result contained the probability of active and inactive compounds symbolized by Pa and Pi. The selection criteria include compounds having higher Pa values than Pi and are treated as biologically promising compounds.

Drug-likeness Profiling

The pkCSM server was used to examine the drug's pharmacokinetic characteristics. The ADMET parameters- absorption, distribution, metabolism, excretion, and toxicity are widely predicted and are an essential step of drug discovery. Drug-like qualities, mainly including intestinal absorption, blood-brain barrier permeability, cytochrome P450 interactions, and total clearances, can also be evaluated to determine physiological behavior. Compounds' SMILES were respectively used to comprehensively understand systemic exposure and stability of compounds in biological systems (Bultum et al., 2022; Pires et al., 2015).

Side Effects Profiling Assessment

The possible adverse side effects were generated using the online server ADVER-PRED (http://www.way2drug.com/adverpred/). The server predicts the probable side effects based on structure similarity trained on known drug safety datasets, thereby facilitating early safety evaluation during drug discovery (Ivanov et al., 2018).

Prediction of Compounds Lethality

The online server GUSAR (l), the way2drug platform, was used to assess the acute rat toxicity. The server predicts values in terms of median Lethal Dose of 50% (LD₅₀) for various administering routes, including oral, intravenous, intraperitoneal, and subcutaneous. The compounds were also classified according to the toxicity classification system (Zakharov et al., 2012; Zhu et al., 2009).

Simulation Studies for Stability Analysis

Based on integrated analysis of multiple computational approaches, the top-performing compounds were selected for stability investigation. Therefore, the simulation studies were conducted to evaluate dynamic behavior and structural ability in a defined system. The platform GROMACS with the version of 5.1.2 was selected at a temperature of 300k. The topologies and protein-ligand complexes files were generated using gmx grep utility (Khan et al., 2022; Saxena et al., 2019). The CGenFF (CHARMm general force field) was then applied for ligand parameterization, while the system was solvated using the TIP3 cubic box model. Subsequently, charge neutralization was performed by adding salt after the boundary conditions (Liu et al., 2025; Vanommeslaeghe et al., 2010). Steepest descent algorithm with 25,000,000 steps was utilized for minimizing the energy. With constant volume of temperature and pressure conditions (NVT and NPT), the temperature was set. As the last step, interactions were computed using the mesh Ewald method, and the production simulation was conducted at 100ns to gain deeper insights. Trajectories such as root mean square deviation and fluctuation, radius of gyration, solvent-accessible surface area, and hydrogen bonds were analyzed later on to obtain and evaluate the refined complexes (Noumi et al., 2025; Petersen, 1995; Rahmati et al., 2025).

RESULTS

Screening of small molecules

From the Mcule platform, docked compounds were generated. Along with the RO5 violations from 0-1, a hundred Mcule molecules were produced. Out of which 17 compounds showed 0 violations, while 44 compounds turned up having 1 violation. The binding energies range from -11.3 kcal/mol to -8.8 kcal/mol and were recorded by the docking investigations, which were greater from the reference compound. The inhibitor compound was taken from the protein itself. The structural toxicity of every single compound was assessed through the Mcule platform’s toxicity checker. A total of 39 compounds were found to be structurally toxic. The safer compounds were stored for further evaluation. A compound with 0 violations, along with the binding energies has been shown (Table 1).

Cytotoxicity Screening

The potential of selected compounds against lung cancer cell lines was evaluated with the CLC-Pred server. Out of all the 17 compounds having 0 violations, all the compounds exhibited potential cell lines. Compound 7 exhibited the highest predicted activity with a greater Pa value against the NCI-H460 NSCLC cell line. Additionally, it also demonstrated broad-spectrum activity across various cell lines, indicating strong anti-cancer potential. On observation, all the compounds predicted various cell lines in lung cancer. Interestingly, the reference compound showed moderate activity across cell lines, with the highest Pa value (0.511) observed only for SK-LU-1 adenocarcinoma cell lines, while the rest of the compounds exhibited broader activities suggesting superior cytotoxicity potential against lung cancer (Table 2).

Assessing Biological Activity Prediction

The biological activity prediction analysis was performed using the online software PASS prediction. The results in Table 3 revealed the compounds showing different probability scores against parameters. Different activities related to lung cancer were targeted. The results were analyzed based on the probability of being active and inactive. All compounds gave favorable results.

Among all, compound 2 demonstrated the highest predicted antineoplastic activity with an active probability score of 0.602, indicating strong anticancer potential. It also exhibited strong angiogenesis inhibition, indicating tumor associated blood vessel inhibition. Interestingly, compound 6 also displayed broad-spectrum activities such as anti-inflammatory, anti-allergic, and anti-asthmatic, indicating its potential major supportive role in lung-related conditions/symptom.

Drug-likeness Evaluation

The lead-like molecule depends on various factors; one of the main factors is its favorable pharmacokinetic properties. The drug kinetic profile of all compounds was assessed using the ADME and pKCSM prediction tool containing several parameters such as water solubility, Caco-2 permeability, human intestinal absorption, skin permeability, AMES etc. the predicted properties of all screened compounds with the reference compounds are given in table (Table 4).

The analysis revealed compounds demonstrated acceptable properties with high intestinal absorption and membrane permeability. HIA ranges from 96% - 85%. The highest Caco₂ permeability recorded was 1.283 for compound 2. Except compound 1, all compounds were AMES negative (non-mutagenic) having a good clearance range. The max dose calculated was 0.5 to 0.06. Overall, the compounds represent a safety window.

Prediction of Safety Concerns

Adver-Pred server was used to assess possible side effects of filtered compounds. The prediction was based on calculating the probability of activity and inactivity of various drug reactions. The lower Pa values compare to Pi indicates a minimal likelihood of adverse effects and improved safer profile. Effects such as arrhythmia, cardiac failure, myocardial infarction, hepatotoxicity and nephrotoxicity. The predicted values of all compounds with the control is mentioned in table (Table 5).

The data revealed the minimum side effects recorded for compound 6 and compound 7 having only 1 side effect. Notably, compound 2 demonstrated moderate effect probabilities with two side effects, with no evidence of severe organ-specific outcomes. Compound 1 exhibited the lowest probability of adverse effect. Comparatively with control, it delivered higher probabilities of cardiotoxic effects. The results indicate that the compounds represent better safety profiles relative to control.

Lethal Dose Evaluation

Acute rat Lethal Dose (LD₅₀) profile was predicted by GUSAR online platform through multiple routes to determine if the compounds fall within the Applicable Domain (AD) to ensure reliability (Gadaleta et al., 2019; Zwickl et al., 2022). Every route was determined in mmol/kg, mg/kg, and with the safety class shown in Table 6.

The table revealed that all the compounds showed higher values, indicating a higher dosage threshold before toxicity occurs. The highest dose record as 1972 mg/kg in the oral route by compound 2, suggesting excellent safety upon oral consumption. Similarly, compound 6 also records the highest value for 1469 mg/kg for oral administration, indicating lower toxicity. Compound 1 represents the highest dose as 1594 mg/kg for SC routes. Interestingly, most of the predictions, including control, fall into classes IV and V, which are characterized as low-toxicity categories, indicating a favorable toxic profile. The findings suggest compounds possess an acceptable window for further investigations.

Based on the integrated analysis of all the parameters analysis top two compounds were selected and identified for stability analysis enrichment, namely, compound 2 and compound 6, due to their balanced profiles for anticancer activities, favorable pharmacokinetics characteristics, and acceptable safety profile. Therefore, the protein-ligand interaction diagram of these two compounds with the reference compound has been illustrated (Figures 1, 2A-2C). Optimizing these compounds enables evaluation of different trajectories, thereby validating the docking results and detecting the most promising therapeutic lead candidate.

Click to enlarge

Click to enlarge

The diagram represents such as van der Waals interaction, conventional hydrogen bonds, Pi alkyl, alkyl, Pi sigma, carbon hydrogen bonds, and halogen interactions. while the reference compounds lack hydrogen bonds. But, on the other hand, the selected compounds shared common residues with the control. Table 7 represents interacting residues.

The commonly residues are observed as PHE428, PHE302, MET379, LEU387, PHE386 and PHE406. However, some additional residues, TYR382 and VAL384 are also observed in compound 02 from the reference inhibitor indicating a strong interacting profile of the compounds which supports the findings.

Molecular Dynamics Analysis

Deviation and Residual Fluctuation Analysis

The deviation profile (RMSD) was analyzed to find out the stability of the structure of protein-ligand complex throughout 100 ns simulation. The analysis indicated that compound 2 and reference compound showed an increment during the initial phase of simulation and stabilized around 1.2-1.5 nm, indicating moderate deviations. In contrast, compound 6 was observed to exhibit lower RMSD values within the range of 0.5-0.8 nm with some minute fluctuations observed after 60ns, suggesting enhanced stability (Figure 3A).

Click to enlarge

Similarly, RMSF analysis was also performed to assess the flexibility of residues and the dynamic behavior of the protein. the noting point observed is that the majority of the residues exhibited fluctuations below the range of 0.5 nm, indicating overall stability. However, the fluctuations observed around 320-400 range might represent a loop of binding-associated regions. The reference inhibitor displayed higher fluctuations approximately at 1.5 nm. whereas compound 2 showed moderate fluctuations, and minimal fluctuations were observed for compound 6 throughout the range (Figure 3B).

Radius of Gyration and SASA Analysis

The Rg was evaluated to determine the compactness and depict the behavior of protein complexes during the simulation period. The reference compound and compound 2 also displayed relatively stable values between 2.8 and 3.2 nm, which suggests maintain structure with slight fluctuations. Conversely, compound 6 demonstrated a lower and highly stable Rg value of approx 1.7-1.8 nm, which indicates compact structural conformation. Moreover, the SASA (solvent accessible surface area) was analyzed to depict the extension of protein surface exposure to the solvent during simulation. Plots of SASA also depict compounds 2, and the reference compound relies on the same manner to maintain values between 45-55 nm². This reflects higher exposure and moderate solubility. Comparatively, compound 6 exhibits better results, ranging from 7-10 nm2, which remain stable throughout the simulation. The plots of both strategies are given in Figures 3C and 3D.

Solvation Free Energy Analysis and Hydrogen Bondings

The solvation profile was evaluated to check which solvent molecules surround and interact with the solute ions. The values are plotted against a 100 ns runtime. Compound 6 stands out alone, depicting higher values of fluctuations around -10 to -20 kJ/mol, reflecting constant energy across the simulation. Meanwhile, both the compounds, reference and compound 2, showed similar stability, fluctuating around -60 and -80 kJ/mol, indicating favorable solvation conditions during simulation. despite the differences, all compounds showed stability over the run time. From the hydrogen bonds interpretation, maximum hydrogens bonds 4-5 were seen from the reference compound at initial stage and gradually decreased further. Whereas, compound 2 bonding around 15 -2 ns showing bonds from 0-8 and reach up to 9 bonds at last stage showing fluctuations. Compound 6 maintains the hydrogen bonding at initial level showing 9 bonds. Afterwards, it also shows bouncing fluctuations between 6-8. Both the compounds showed much better results from the reference compound. The graphical plots of SASA and hydrogen bonds have shown in diagram (Figures 3E and 3F).

DISCUSSION

The present study investigated the therapeutic screen of the TEAD2 palmitoylation pocket targeting lung carcinoma by utilizing a computational drug discovery approach. In the current scenario, TEAD2 emerged as a prominent regulator of the Hippo signaling pathway, which heavily drives tumor growth when activated, leading to therapeutic resistance in lung cancer. since palmitoylation plays a primary role and is vital for maintaining TEAD2 conformational stability and interaction with co-activator proteins such as YAP/TAZ (Kumar et al., 2024). In view of that, inhibition of the binding pocket of TEAD2 is considered a promising path for intervention as palmitoylation is essential for the stability of TEAD2. Therefore, the current research study focused on structural based virtual screening through the Mcule platform, an online server (Ott et al., 2025; Sun et al., 2022).

The platform enables the screening of small molecule inhibitors with strong binding energies comparatively to control. The control HUY itself attached to the reference protein PDB:6E5G was selected. The identification of promising therapeutic drug candidates required evaluation of multiple approaches to ensure effectiveness and safety. Therefore, integrated in silico approach involving cytotoxic prediction, biological activities, ADMET profiling and acute rat toxicity with boost of MD was employed. Initially, Mcule platform revealed 7 compounds having 0 violations having the best energy were proceed with cell lines prediction. Various cell lines had been predicted for all compounds. compound 2 (MCULE-4863298419-0-1) demonstrated moderate cytotoxicity against NSCLC cells suggesting distinct lung cancer subtypes. Similarly, compound 6 (MCULE2504052738-0-1) exhibited cell lines such as COR-L23, a large lung cell carcinoma, NCI-H1985, a bronchio carcinoma, and NCI-H1993 NSCLC, representing a wide spectrum of cell lines. Moreover, validation through the PASS parameter provides insights of compound 2, demonstrating the strongest antineoplastic activity, higher than control, depicting support to suppress tumor-associated blood vessel formation, which is a mechanism involved in invasion and metastasis, along with growth of cells. Angiogenesis is highly considered as proliferating tumors found in Compound 2 (C2). Whereas, Compound 6 (C6) also demonstrated multiple activities, such as anti-asthmatic (Pa=0.751), antiallergic (Pa=0.720), and anti-inflammatory (Pa=0.458). These activities are particularly present in lung cancer as pulmonary malignancies are usually associated with inflammation, respiratory distress, and altered immune response. On the other hand, drug-likeness also revealed favorable results. C2 was predicted to be non-mutagenic and showed no hERG-I inhibition, suggesting a lower risk, possessing superior characteristics desirable for oral administration with safety. While C6 also exhibited supporting results and maintained the overall pharmacokinetic profile. Compared to the control, both compounds stand out. AdverPred analysis and side effect profiling further pointed out the importance of efficacy with toxicity. Although C1 demonstrated fewer side effects but lacked other parameters, thereby reducing its potential. In contrast, C2 and C6 demonstrated manageable toxic profiles while standing out among other parameters as well. Interestingly, both compounds also strengthen the LD₅₀ analysis, exhibiting the highest value 1972 mg/kg by C2, enhancing safety upon oral administration, and 1469 mg/kg by C6, suggesting low toxicity. Both compounds exhibited higher doses than the control. The dynamics studies performed for these top two compounds further revealed as a support. Different trajectories, such as RMSD, RMSF, SASA, Rg, and hydrogen bonds, were conducted, indicating stability, compactness, and interactions of the complexes. From the dynamics analysis, the results demonstrated MCULE2504052738-0-1 (C6) exhibited the highest structural stability among the tested compounds, showing lower RMSD values, minimal fluctuation of residues, stability, and compactness throughout the simulation. However, other compounds show a similar range, with the reference compounds also exhibiting good results. Furthermore, maximum hydrogen bonds were recorded for both compounds 8-9 bonds, compared to the control, which only showed 4 bonds at the starting phase. Overall, the findings support the selection of both compounds, indicating therapeutic evidence.

CONCLUSION

In the present study, an integrated in silico multi-approach was conducted to identify the promising lead candidate against lung carcinoma. The findings revealed the screened compounds exhibiting varying degrees of anti-cancer potential, drug-likeness profile, safety profiling, and acute toxicity profile, emphasizing the vital role of multiple strategies for rational candidate selection. Additionally, the molecular dynamics studies were also conducted to support the evidence. After investigations, among the investigated compounds, MCULE2504052738-0-1 and MCULE-4863298419-0-1 were proven best for all the parameters. Overall findings indicate that MCULE2504052738-0-1 emerged as the most promising lead candidate, demonstrating a superior balance profile of therapeutic efficacy, pharmacological relevance, and safety with higher stability and robustness. However, to confirm its advanced therapeutic potentials, experimental validation is required.

ACKNOWLEDGEMENT

The author extends the appreciation to the Deanship of Postgraduate Studies and Scientific Research at Majmaah University for funding this research work through the project number (R-2026-399).

ACKREVIATIONS

TEAD

Transcriptional Enhanced Associate Domain

NSCLC

Non-Small Cell Lung Cancer

SCLC

Small Cell Lung Cancer

YAP

Yes-Associated Protein

TAZ

Transcriptional Co-Activator with PDZ-Binding Motif

PDB

Protein Data Bank

LD₅₀

Lethal Dose of 50%

RMSD

Root Mean Square Deviation

RMSF

Root Mean Square Fluctuation

PASS

Prediction of Activity Spectra for Substances

MCULE

Molecular Computational Utility for Lead Exploration Database

ADMET

Absorption, Distribution, Metabolism, Excretion, and Toxicity

CLC-Pred server

Cell Line Cytotoxicity Predictor Server

SMILES

Simplified Molecular Input Line Entry System

pkCSM server

Pharmacokinetics of Small Molecules server

ADVER-PRED

Adverse Effect Prediction Server

GUSAR

General Unrestricted Structure Activity Relationships

GROMACS

GROningen MAchine for Chemical Simulations

CGenFF

CHARMM General Force Field

CHARMm

Chemistry at HARvard Macromolecular Mechanics

TIP3

Transferable Intermolecular Potential with 3 Points

Rg

Radius of Gyration

SASA

Solvent Accessible Surface Area Analysis.

CONFLICT OF INTEREST

The authors declared that there is no conflict of interest.

FUNDING

The investigation contains no funding from any source.

ETHICAL APPROVAL

Not applicable.

AUTHOR CONTRIBUTIONS

Amani Alharthi: Conceptualization, experimental design, data curation and analysis, literature review, manuscript drafting and final manuscript approval.

GENERATIVE AI STATEMENT

Language editing and refinement was done using only generative AI tools (e.g., Grammarly and QuillBot). The author is entirely responsible for the content, accuracy and integrity of the manuscript.