Cheminformatics and Pharmacological Network Analysis for the Prediction of Molecular Mechanism of Herbal Drug Ligands


 The global demand for safer and more therapeutically effective medications is surging, providing medicinal plants a boost as suppliers of lead particles. The focus of current research is on an in silico comparison of one major bioactive principle and curatively designed new small drug-like molecule (scaffold analog). The recent study confirmed that the plant belongs to the Cyperaceae family and that it is Cyperus rotundus L. in taxonomy. The study's purpose was to uncover the mechanism of action of ligands/ scaffold analog by revealing genomic relationships, cellular signaling, and the top ten diseases/ illnesses that they were linked to. The scaffold analog showed promising drug-like potential as compared with cyperene. These investigations could broaden the reach of herbal medications, provide new formulations for current diseases or disorders, and pave the door for drug repurposing.


Introduction
The present work signi cantly emphasizes the curatively designed new small drug-like molecule (Scaffold analog) and in silico performance of comparative studies with a naturally found bioactive constituent (Cyperene) of the plant through ligand based approach (chemical similarity criteria) to unveil prospective target sites, genomic interactions, cellular signaling pathways, and their associated diseases/disorders through proteogenomic analysis to better understand their mode of action. Herbal medicine has been used by humans for healing purposes in many treatment systems (Ayurveda, Unani, Homeopathy, and Siddha) from ancient times, and has been described as an essential source of medicine even in the present period [1]. Obesity, remittent fever, monthly irregularities, bowel disorders, colitis, diarrhea, vomiting, uterine contraction, bad breath, and other neuromuscular-related issues are all treated by Nut Sedge (Cyperus rotundus L.) [2,3]. The major therapeutic agents are found in rhizomes of the herb. The plethora of bioactive constituents is found in it including cyperene, α-cyperone, humulin, βselinene, α-selinene, cyperotundone, zierone, α-calacorene, campholenic aldehyde, pinene, γ-muurolene, longiverbenone, β-caryophyllene oxide, and limonene [4]which represents different classes of secondary metabolites such as alkaloids, phenols, tannins, saponins, steroids, coumarins, avonoids diterpenoids, and triterpenoids [5,6]. Herbal-derived ligands and targeted therapeutic approaches may be successfully worked on using in silico approaches [7]. The ligands (bioactive component, scaffold analog) of Cyperus rotundus L. were evaluated to nd its molecular activities at different target sites by maintaining a curative space for drug discovery, drug reuse[8], development, advancement, and innovative results [9]. These efforts include an in silico approach to cheminformatics [10], and proteogenomic analysis [11] including physicochemical analysis, pharmacokinetics, pharmacodynamics, medicinal chemistry, drug similarities, designing of novel scaffolding analog, target predictions, molecular docking, identi cation of genes, genomic interactions, cellular signaling, and their biological insights (top ten diseases/ disorders regulated by these interactive genes via identifying the target sites of ligands) through representative overexpression analysis (ORA) [12], and Network Topology-based Analysis (NTA) [13]. The topological polar surface area (TPSA) of a molecule is de ned as the surface sum of all polar atoms or molecules including oxygen, nitrogen, and their attached hydrogen atoms [5]. Log P is the logarithm of the partition coe cient (P) when one of the solvents is water and the other is a non-polar solvent. For the optimal results, the log P value is 0<LogP<3, LogP<0: poor lipid bilayer permeability, and LogP>3: poor aqueous solubility. It is a measure of the lipophilicity or hydrophobicity of the compound. XLogP3 predicts the octanol/water partition coe cient of an organic compound. It is based on the group contribution method with suitable corrections. Water solubility (log S) is represented from insoluble <-10<poorly<-6<moderately<-4<soluble<-2very<0<highly soluble. The pharmacokinetics parameters include the absorption (gastrointestinal absorption, p-glycoprotein substrate), distribution (blood-brain barrier), metabolism (CYP1A2 inhibitor, CYP2C19 inhibitor, CYP2C9 inhibitor, CYP2D6 inhibitor, and CYP3A4 inhibitor), elimination (elimination half time (T 1/2 ):3h<T 1/2 <8h and clearance (cl): 5ml/min/kg<cl<15ml/min/kg, and toxicity (human ether-a-go-go-related gene(hERG-category 0: non blocker; category 1: blocker), human hepatotoxicity (H-HT-category 0: negative; category 1: positive), Ames mutagenicity-category 0: negative; category 1: positive, skin sensitization-category 0: non sensitizer; category 1: sensitizer, drug-induced liver injuries (DILI-category 0: negative; category 1: positive), and FDA maximum recommended daily dose-category 0: FDAMDD negative; category 1: FDAMDD positive). The bioactivity scores of the ligands on different target sites such as GPCR ligand, ion channel modulator, kinase inhibitor, nuclear receptor ligand, protease inhibitor, and enzyme inhibitor activities are covered under pharmacodynamics. The bioactivity of screened molecules may then be calculated as a sum of activity contributions of fragments in these molecules. This provides a molecule activity score (a number, typically between -3 and 3). Molecules with the highest activity score have the highest probability to be active. The drug-likeness properties were evaluated by the application of 'Lipinski Rule of Five', 'Ghose's Rule', 'Veber's Rule', 'Muegge's Rule', and 'Egan Rule'[14-16]. Pan-assay interference structure (PAINS) analysis is performed on compounds having desirable physicochemical properties to determine their toxicity. This assay is also called toxicophores because of the presence of some group elements that disrupt the biological processes through interference with DNA or protein which results in fatal conditions such as carcinogenicity and hepatotoxicity. Brenk analysis gives an idea about structural alert including chirality and steric hindrances. The lead-likeness comprises the ligands having 250≤MW≤350, XLogP≤3.5, and no. of rotatable bonds≤7. The synthetic accessibility score from 1 (very easy) to 10 (very di cult) to synthesize.

Physicochemical, Pharmacokinetics, Pharmacodynamics, Drug-Likeness, and Medicinal chemistry attributes of the ligands
The cheminformatics of the bioactive constituent (Cyperene) and a Scaffold analog are given in Table 1. The IUPAC nomenclatures were used for further generation of smile notations and .mol2 les[31] for further exploration. A scaffold analog (SA) was drawn by taking into consideration optimum cheminformatics parameters. Table 1 shows the ligands' IUPAC names, smile notation, and twodimensional structure. The molecular weight, hydrogen bond acceptor (HBA), hydrogen bond donors (HBD), lipophilicity (LogP, XLogP3), water solubility (Log S), and topological surface area are all shown in Table 2. The scaffold analog shows prominent and favourable properties as compared to cyperene. Table 1 Showing ligand name, IUPAC name, and 2-D structure The cheminformatics data generates signi cant details about the ligands which have been widely used to screen out potential compounds with desirable properties. The characterization of ligands architecture for drug-like characteristics has further explored the pathways for curated drugs formulations and promising drug repurposing compounds. Table 3 shows the ligands' pharmacokinetics including absorption (GI absorption, P-glycoprotein substrate), distribution (blood-brain barrier), metabolism, elimination (T 1/2 , clearance), and toxicity pro les. The scaffold analog shows good results within the acceptable range of all the stated parameters as compared with cyperene. The pharmacodynamics (bioactivity score at different targets), drug-likeness, and medicinal chemistry attributes of the ligands are given in Table 4. The scaffold analog showed most promising attributes without any violations and alerts as compared to cyperene that indicates that there is fair possibility of developing new drug-likeness molecules from bioactive constituents of the other medicinal plants also in future.

Drug targets and molecular docking
When compared to Cyperene, a key bioactive constituent of Cyperus rotundus L., the Scaffold analogue, a novel and innovative product, showed promising physicochemical, pharmacokinetics, pharmacodynamics, drug-likeness, and medicinal chemistry pro les. SWISS Target Prediction Software assessed both ligands (Cyperene and SA) and projected target sites based on probability as shown in the gure 1 and 2 respectively.
The .mol and .mol2 les of the SA were not available on any database because it was a newly drawn small molecule. The .mol2 le was created and then veri ed, viewed, and standardized by UCSF Chimera.
The .mol les of Cyperene were available on Zinc 15 and PubChem databases, which were further converted to .mol2 les through Open Babel [35] or CORINA Software [36]. As indicated in Table 5, the decision to choose target sites, classes, and PDB ids was based on the outcomes of Swiss target prediction, bioactivity scores at speci c target sites, and ligand compatibility with PDB. PDB les and ligands were standardized and selected for further analysis. In the PDB les, all the missing parameters such as atoms, missing loops, side chains, and residues were checked and inserted. The incorrect chirality, steric clashes, water molecules, and non-protein residues were removed via structure optimization and energy minimization using chimera, and Chiron energy minimization and re nement tool. The structure of the ligand was carefully inspected to make sure that the topology of the molecule is correct, as well as its protonation state and tautomeric form. The binding conformation aids to reveal the binding energy of the ligands with target sites. The visualization of biomolecules illustrates the dynamic aspects of molecular simulation docking and its application in structural biology. The docking helped us to know the interaction of ligands and proteins at the atomic level to signi cantly characterize the behavior of the ligands while interacting with target proteins.
The binding free energy (ΔG bind ) comprises Vander Waals energy (ΔG vdw ), the sum of electrostatic energy (ΔG elect ), the sum of hydrogen bond and desolvation energy (ΔG Hbond ), the sum of nal total internal energy (ΔG conform ), the sum of torsional free energy (ΔG tor ), and the sum of unbound system energy (ΔG solv ). The energies of cyperene and scaffold analog (SA) are shown in Table 6. The Swiss Docking is based on an algorithm that consists of several steps such as a large number of bonding modes (BMs) are generated, either in the user-de ned box (local docking) or near to the target cavities of the whole protein surface (blind docking). Concurrently, their CHARMM energies are evaluated. Then BMs with the most favorable energies are ranked, taking into consideration the solvation model. The results of binding free energies (DIGibind) and compatibility of 3-D crystalline protein structures with ligands are quite important for better elucidating the biochemical pathways. However, the present research discerns the ligand based approach wherein the ligand/ scaffold analog showed activities on multiple targets. The purpose of molecular docking herein is just to establish the binding energy of scaffold analog irrespective of binding a nity. That generated the primitive idea about binding activity.
After swiss target prediction, Cyperene was docked to Peroxisome proliferator-activated receptor alpha (PPARA) a target site of nuclear receptor but it could not be materialized in the case of scaffold analog (SA) due to non-compatibility of enzyme inhibitor (0.27) and nuclear receptor (0.07) target sites. The molecular docking was done on Leukotriene A4 hydrolase, a target site of protease inhibitor which shows a bioactivity score of 0.03. Hence, it became crucial to conduct proteogenomic analysis to have better insights into the genomic interactions and molecular mechanism of action of the ligand/ SA. Over-representation analysis is also a sequential pathway that is performed to nd out the cellular signaling pathways and most promising (top ten) diseases/ disorders controlled by these interactive genes.
(ii) The sequential pathways for analyzing interactive genes and their associated diseases/disorders:-ORA>Disease>Disgenet>genome>Submit>Result 3.3.1. Cyperene Interaction grid and signaling pathways: The cyperene may prove as a potential remedy for enlarged polycystic ovaries, prostatic intraepithelial neoplasia, female infertility, marijuana abuse, insulin resistance, mental depression, obesity, mood disorders, depressive disorders, and schizophrenia.
3.3.2. Scaffold Analog interaction grid and signaling pathway: The novel scaffold analog may prove as a remedy for anuria, Allanson pantzar McLeod syndrome (a genetic disorder), endogenous depression (melancholia), coronary restenosis, widely patent fontanelles, and sutures, nephrosis, Behcet's syndrome (autoimmune disorder), atherosclerosis, cerebrovascular accident, and heart failure. The prominence of herbals is ostensibly observed among comestible practices as well as in holistic healthcare practices embraced by society [6]. The global demand for prospective lead molecules is growing, and active medication particles have been diverted through the use of in silico techniques to speedy health problem resolution [7]. There is always a demand for safer and more effective therapeutic compounds, which can be answered by doing Cheminformatics and Proteogenomic studies on indigenous herbal medicine connections. In the present study ligand-target approach of the investigation was performed to elucidate the pathways, interaction of genes involved in speci c cellular pathways, and controlling the speci c diseases, and disorders. The ligands and their generalized mode of action affect genes and are responsible for which diseases/ disorders. The conventional and CAM (complementary and alternative) is a promising strategy for ghting against cancer. There is a huge demand for newly discovered phytochemicals for performing preclinical and epidemiological studies to identify their molecular signatures[8]. There is a fair possibility of identi cation of more potent herbal analogs for curing arthritis, diabetes, and cancer through the application of in silico approach [9]. Although a myriad of literature is available on target-based approach and speci c target sites were investigated for the different ligands which did not pave the way for exploring new ligands and their scope for drug repurposing is left behind. Hence, the present investigation worked on a ligand-based method (chemical similarity criteria), which does not guarantee similar bioactivity. For this very purpose bioactivity score at a different target, sites are evaluated, and their binding energy is measured by swiss docking. Although, the objective of docking was not extra precision but to just evaluate the binding energy at a particular class of receptors as estimated by swiss target prediction. The medicinal plants are the least studied facet on this point of view and the scope of innovative drug designing from their active chemical constituents through the application of curative and synthetic pathways is yet to be achieved and untapped. The present discerns that the least investigated and explored aspects of in silico drug designing through innovatively and curatively fabricating drug-like small molecules. Further, the speci c molecular docking studies comprising interaction analysis (protein-ligand), replica exchange molecular dynamics (REMD), MM-GBSA, and APBS electro statistics which is validated by Ramachandran torsion plots, this approach may be applied to calculate forces, energies, and binding a nity [9]. These scienti c efforts will not only improve the biological e cacy of the current drug but also provide a roadmap for innovative drug development or a fair potential for drug reuse after pathways are revealed at the genomic level [10]. Furthermore, it may be assumed that there is a wide scope for investigation of other members of the Cyperaceae family including Scirpus grossus L., Trichophorum clintonii (Gray) S.G.., and Eleocharis dulcis (Burm.f.) Trin. ex. Hensch. etc. for the development of herbal medicines.

Conclusions
The current study uses a ligand-based strategy (chemical similarity criteria) to identify the therapeutically developed small drug-like molecule (SA) and compares it to a naturally occurring bioactive ingredient, Cyperene. The bioactivity scores derived using molinspiration cheminformatics provided activity pro les of the ligands at several target locations, and they were subsequently swiss docked based on their highest activity ratings at each site. It sparked some notions regarding their binding energies. The Swiss target prediction created ideas for potential ligand targets, which were then entered into the webGestalt web server software to determine their relationships, signaling cascades, and diseases/ illnesses connected with them. The studies' primary goal is to gure out where ligands might be able to target, how genes connect, how signaling pathways work, and how they relate to the top ten diseases and disorders.
This would open the door to more medication development and possible drug repurposing compounds. In a broad sense, the current investigation resulted in the examination of many ligand-target locations as well as their potential mode of action. These ndings can then be con rmed by testing the activity of target-speci c ligands using molecular mechanics generalized Born surface area continuum solvation (MM-GBSA) to determine their resilience and electrostatic stability. However, this is outside the scope of the current study.
Researchers are pursuing a complete understanding of the herbal therapeutic agent's mechanism of action to demonstrate its potential as a drug for drug reuse and design successor due to its widespread availability in nature and long-term use in a variety of diseases. An in silico technique was used to analyze Cyperus rotundus L., a member of the Cyperaceae family. Cyperene, a major bioactive ingredient of the herbal medication, was examined along with the scaffold analog. The cell signaling, regulating, and regulatory pathways were better understood, thanks to cheminformatics and proteogenomic research. The scaffold analog has better drug similarity and safety characteristics than the originals. The main potential of indigenous medications remains untapped, which may be scienti cally researched to meet global demand for pharmaceuticals, according to the ndings. Based on this extensive research, it can be deduced that the scaffold analog and the bioactive components, either together or separately, could be promising effective molecules for phytomedicines in additional disorders, as mentioned in the pharmacological analysis.

Con ict of interest
The authors report no con icts of interest. The authors are responsible for the content, and writing of this article.

Acknowledgements
Thanks to Honorable Dr. Chinmay Pandya (Pro-Vice-Chancellor, Dev Sanskriti Vishwavidyalaya) for his kind, un inching and unwavering motivation. Further, we are obliged to the experts of Medicinal Plants

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.