Hepatitis C, the silent disease caused by Hepatitis C virus (HCV) is a chronic health infection globally. HCV causes permanent hepatic cirrhosis and carcinoma in humans. WHO estimated about 3 million incidents of HCV infection around the world. Multiple variant genotypes along with the development of Quasi-species limited the efficacy of drugs used for the treatment of HCV infections. This heterogeneity of the virus hampered the drug development against them. The virus hosts many structural and non-structural (NS) proteins. NS5B is a non-structural protein with a unique structure and function. The protein is a RNA dependent RNA polymerase (RdRp) responsible for building the vital genetic component of the virus. Inhibition of NS5B stops viral replication and propagation. The major role played by RdRP makes it a preferential target for anti-HCV drug development. An association of docking and rescoring studies was performed to 24 compounds derived from various plant sources to estimate their activity against HCV NS5B RdRp. Based on the docking characterization and ADMET properties andrographolide, esculetin, columbin and tinosporide were identified as they showed greater potency against HCV NS5B RdRp. However, based on hepato bioactive spectrum and ADMET score, andrographolide from Andrographuis paniculata emerged as a strong contender with lead like characteristics acting as a promising drug candidate.
Non-structural protein, RNA dependent RNA polymerase, docking, Andrographolide.
Hepatitis C virus infections represent primary public health concern. It is estimated that 3 to 4 million people are chronically infected globally. HCV infection increases the risk of liver cirrhosis and hepatocellular carcinoma development . HCV infected individuals are treated with interferon, Ribavirin (a nucleoside analog) combination and other approved protease inhibitors . Hepatitis C is termed as a silent disease since the incubation period of HCV ranges from 6 to 8 months with no symptoms, making early detection difficult . However, the therapies have limited efficacy against HCV genotypes. HCV has been classified into different strains based on their genetic differences. HCV diverges into 6 genetic variants. Genetic diversity is also reported within the subgroups making genetic heterogeneity responsible for differences in disease outcome and response to treatment among HCV infected individuals . Severe drug regime, side effects and high treatment costs limit patients’ compliance with the treatment. Therefore, a major focus towards development of cost effective and efficient anti-viral therapeutics targeting viral proteins with limited effects on the host system is of high priority.
HCV contains 9.6 kb positive – sense RNA genome and belong to Flaviviridae family. The viral polyprotein is processed into individual structural proteins – Core, E1, E2 and P7 and nonstructural(NS) proteins – NS2, NS3, NS4A, NS4B, NS5A and NS5B . Host cytoplasm acts as a factory for HCV replication. RNA dependent RNA polymerase (RdRp) activity of HCV nonstructural protein NS5B mediates viral RNA genome replication. The importance of NS5B in synthesizing HCV RNA makes it a preferential target for anti-HCV inhibitor development. Furthermore, the host cells lack RdRp activity leading to minimal or no side effect on the host counterpart.
HCV NS5B has a number of enzymatic and structural differences from cellular RNA polymerases, which make it a promising target for small molecule antiviral development . Phytochemicals from medicinal plants shows prominent role in prevention and therapy of many diseases . Our research focuses on the insilico identification of promising bioactive molecules against HCV infections by considering NS5B as a primary target.
Materials and Methods
Identification of protein structure and preparation of small molecules
The NS5B is 65 kDa proteins with a typical ‘right hand’ polymerase shape containing finger, palm and thumb domain. The lambda 1 loop of HCV NS5B, which extends from the fingers to the thumb domain, is unique to RNA-dependent RNA polymerases and the beta-hairpin toward the catalytic region is specific for HCV NS5B. The structure of NS5B encloses motifs A-F, which are crucial for their functional activity . Brookhaven protein data bank (PDB) is a major resource repository of three-dimensional structural information of biological macromolecules like nucleic acids and proteins. The X-ray structure of NS5B protein was retrieved from PDB (PDBID- 2YOJ) .
Traditional plant derived medicines are proved to be efficient in the treatment of liver disorders . The bioactive compounds which can act against HCV NS5B were collected from literature [7, 11] and books on Indian medicine (Table 1). The SMILES (Simplified Molecular-Input Line-Entry System) notation of the compounds were retrieved from PubChem  and converted to PDB format using CORINA server . The 2D structures of the compounds are presented in the S. Fig. 1.
List of phytochemicals used in the study
List of phytochemicals used in the study
Binding site prediction – motif conservation
The full genome sequences of eighteen HCV strains belonging to six approved HCV genotypes were retrieved from NCBI  (Table 2). Considering the NS5B sequence of 2YOJ.pdb as the reference sequence all other sequences were multiply aligned to identify the conservation of functionally significant motifs A-D using CLC workbench (www.clcbio.com). The identified conserved motifs were considered as the ligand binding site.
Genbank accession numbers of complete protein sequence of HCV genome
GB accession number
GB accession number
Genbank accession numbers of complete protein sequence of HCV genome
Molecular docking and ADMET predictions
Molecular docking predicts preferred orientation of the ligand molecule to the protein when bound together to form a stable complex. The scoring function gives the strength of association between them. AutoDock 4.2  was used for docking study with the Lamarkian genetic algorithm to find globally optimized conformation. The grid spacing was set to 0.403 nm and grid box dimension was set to 86 x 60 x 60 which enclosed the residues of motifs A to D. Default settings were applied for remaining parameters. At the end of a docking with multiple runs, a cluster analysis was performed. Docking solutions were clustered together and ranked by the lowest docking energy. The lowest binding energy cluster was selected as the representative binding mode. Out of twenty four bioactive compounds, ten compounds with least binding energy were considered for the next phase of screening.
Absorption, distribution, metabolism, excretion and toxicity (ADME/T) properties receive more concern in rational drug design, as they determine the development of safe orally bioavailable drug. The determination of characteristics of compounds that are more likely to exhibit satisfying ADME/T properties has led to the concept of “druglikeliness”. Drug likeliness properties obeying rule of thumb  and molecular properties governing pharmacokinetics of the drugs, including human intestinal absorption, aqueous solubility, plasma protein binding and Ames mutagenicity values  were predicted using MolInspiration  and ChemSilico server (www.chemsilico.com). These parameters act as a filter to screen out more potential leads.
Biological activity profiles are precise indicators not only for molecular properties, but also for biological response of the molecules. Biospectrum connects chemical scaffolds with biological activity. PASS (prediction of activity spectra for biologically active substances) server  predicts the pharmacological effects and biochemical mechanisms of biologically active substances based on their structural formula. The biological activity spectrums for compounds emerging out of screening processes were predicted. The pass activity score for hepato-protectant activity (HPA), hepatic disorder treatment (HPT) and antiinflammatory activity (AIA) were taken under consideration for further analysis.
Molecular dynamics simulation
The molecular dynamics simulation (MDS) calculates the time dependent behavior of the molecular system. They are used to investigate the thermodynamics of biological macromolecules and their complexes. MDS was performed using GROMACS 4.5.3 . Gromos96 forcefield  was used to prepare the protein topology file. Ligand topology and forcefield parameter file were prepared using PRODRG server . The full system was subjected to 5 ns MDS at 300 K temperature and 1 bar pressure. The best protein ligand complex was subjected for molecular dynamics simulation study.
Results and Discussion
NS5B structural and functional analysis
The hydrophilic nature of NS5B was predicted as – 0.201 based on a Gravy index (grand average of hydropathy) by Expasy ProtParam . RdRp (Fig.1) is a compact globular protein divided into finger, palm and thumb domains. The finger domain has two sub-domains, α finger with seven α helix and β finger with four β strands. The β finger is connected to thumb domain through two loops Ala9-Thr41 (loop 1) and Asn142-Ala157 (loop2). The incoming ribonucleoside tri-phosphates (rNTP) enter through a small hole at the bottom of loop 2 to reach the binding cleft. The palm domain is a combination of three anti-parallel β strands and four helices . It forms the most important catalytic center of RdRP and contains highly conserved motifs A to F. Seven α helices and an anti-parallel β strand constitutes the thumb region. The processed RNA duplex is released with the help of the β sheet in the thumb domain .
The multiple sequence alignment of NS5B protein sequences from six genotypes [43, 44] and their sub-types approved by ICTV (International Committee for the Taxonomy of Viruses) were proved to be conserved in RNA dependent polymerases by MSA. Motif A (Asp220-Asp225), motif C (Gly317-Asp319) and motif D (Ala342- Tyr346) are involved in nucleotide tri-phosphate binding (NTP) and catalysis while motif B (Ser282- Asn291) is involved in template-primer positioning. The conserved patterns of motifs A-D are highlighted in Fig.2. The residues involved in divalent cation binding, substrate binding, template selection and differentiation remain highly conserved thus validating the binding site preferences of the drugs.
To identify novel candidate inhibitors of HCV polymerase protein, the twenty four compounds were docked in the catalytic center of the palm region. Different binding poses of twenty four compounds were searched and ranked based on their binding affinity. Top ten compounds (andrographolide, esculetin, tinosporide, columbin, kutokoside, arjunic acid, chasmanthin, silymarin, gossypin and hypercin) were identified based on the AutoDock ranking (Fig.3). The binding energy distribution of the first ten compounds ranged from -5.42 kcal/mol to -10.43 Kcal/mol. The Binding free energy and binding interaction patterns of all the ligands were analyzed using Discovery Studio Visualizer and PyMol. The free binding energy and interaction patterns of the ligands represented in Table 3 are considered as the basic criteria for efficient activity. The binding mode analysis of the compounds with NS5B provides a better insight into the essential residue interactions.
Cation-п interactions  are strong, non-covalent binding force conferring stability in ligand-receptor complex. Strong cation-п interaction was reported among the aromatic group of the ligands with the residues Arg158, Lys141. The lower end of the β finger of NS5B protein forms a rim of basic amino acids (Arg48, Lys51, Lys141, Lys155 and Arg158) to anchor negatively charged rNTPs and Arg158 plays a key role in rNTP binding. The blockage of this region can inhibit the RdRp activity of the protein. HCV NS5B shows a strong preference towards RNA as a template and rNTPs as substrate. Asp225 of motif A and Ser288 of motif B are substrate discriminating residues. It should be noted that, many compounds under our analysis form hydrogen bonds and electrostatic interactions with the above residues, silencing their pivotal function. A hydrophobic pocket formed by Asp225 of motif A and Ser282, Thr281 and Asn291 of motif B accommodates ribose moiety of rNTP. Interaction with these residues further hinders the protein’s functionality. Majority of the ligands interactions were observed with the catalytic aspartate residues Asp220, Asp318 and Asp319. The C-terminal region constitutes the regulatory part of HCV NS5B protein. Ser556 and Ser543 are phosphorylated via casein kinase-2. Upon phosphorylation RdRp becomes more active. Interaction of the ligands with any one of these residue can inactivate RdRp functional propagation. .
Binding energy and interacting residues of all the 24 ligands
Binding Energy (Kcal/mol)
Residues involved in interactions
Thr287, Asn291, Asp318
Ser288, Asn291, Gly317(?-?), Asp318, Tyr555
Lys141, Arg158(?-?), Asn291
Lys141, Asn291, Asp318
Arg158, Tyr219, Thr221, Asp225, Asn291
Lys141, Asp225, Asn291, Asp318
Tyr219, Asp319, Leu320, Ser556
Arg158(+-?), Cys223, Asp225, Asn291, Asn316, Asp318, Asp319, Ser367
Lys141(+-?), Arg158, Ser282, Asp318
Asp318, Asp319, Ser556
Arg158 (+-?), Ser226, Asp318
Arg158(+-?), Asp225, Ser282, Asn291, Asp318
Asp225, Asp318, Cys366, Ser556
Arg158(+-?), Asp225, Ser556
Arg158(+-?), Asp225, Asp318, Asn316
Arg158 (+-?), Asn291
Lys141, Arg158, Ser556
Binding energy and interacting residues of all the 24 ligands
Physiochemical property profile and Bioactivity predictions
The subsequent screening phase was based on physiochemical properties of the compounds under study. The ten phytochemical compounds were evaluated against thirteen physiochemical properties which form the backbone of drug designing process. Compounds which satisfy the Lipinski’s rule of five (RO5) and other physiochemical parameters were selected for further analysis.
Hypericin and kutokoside with molecular weight 504.45 Da and 512.46 Da were excluded from the study, as high molecular weight affects drug solubility, absorption and diffusion. Moderate ClogP (Partition coefficient) is desired for GI track absorption . Based on RO5 compounds possessing hydrogen bond donor <5, hydrogen bond acceptor <10 and total hydrogen bonds <=12 were retained for further analysis . Compounds with CSLogD7.4 (Distribution coefficient) in the range of 1-3 shows moderate solubility and permeability along with favorable in-vivo oral absorption and blood brain barrier (BBB) penetration . Gossypin with the distribution coefficient value of -1.5 was excluded from further analysis as it failed to fall within desired range. Chasmanthin with high intrinsic solubility value makes it an unfavorable lead compound . The intact drug binding to plasma protein affects its displacement in the body  and silymarin with 98% of plasma binding was eliminated from the study. The molar refractivity value of arjunic acid was found to be 187.72 making it unfavorable for entering the next phase . Polar surface area (PSA) measures drug’s ability to permeate cell [49, 50] and its cutoff were between 60 Å2 and 140Å2 for good permeability. Compound kutokoside showed higher PSA value along with high molecular weight. Aromatic and heteroaromatic rings are ubiquitous features in small-molecule drugs. The rings possess fewer degrees of freedom compared to chains; hence show increased drug-receptor binding energy. Less number of aromatic rings (< 3) favors an oral drug candidate . Hypericin with 6 aromatic rings and high molecular weight was excluded from the screening process.
Improvisation of our previous study on phytochemical screening against NS3 protein  was accomplished by performing human intestinal absorption (HIA) and Ames test for crucial selection of phytochemicals against our current target HCV NS5B. Final screening tested HIA  and Ames mutagenecity. The compounds which showed positive HIA and Non-Toxic Ames test were finally selected. After a series of crucial screening tinosporide, columbin, andrographolide and esculetin emerged as promising drug candidates satisfying all the screening criteria (Table 4).
Physiochemical and pharmacokinetic properties of final hits
Physiochemical and pharmacokinetic properties of final hits
C*: Compounds; C1a: Andrographolide; C2b: Esculetin; C3c : Tinosporide; C4d: Columbin;
P1a: CSLogP-Partition co-efficient; P2b: CSLogD7.4-Distribution co-efficient at pH 7.4; P3c : CSLogSo-aqueous solubility; P4d:Plasma binding (%);P5e: Molar refractivity; P6f : Number of atoms; P7g: Polar surface Area,Å2; P8h: Molecular Weight (Daltons); P9i : Number of hydrogen bond acceptor; P10j : Number of hydrogen bond donor; P11k: Number of rotatable bonds; P12l : Number of rings ; P13m: Human intestinal absorption; P14n: AMES test; NT: non toxic
Bioactivity prediction forms the final phase of screening. The Hepato-protectant, hepatic disorder treatment and anti-inflammatory activity were predicted for the final four hits using the Pass server (Fig.4). The compound with Pa (probability to be active) > 0.7 was considered to exhibit the activity in an experiment by the PASS server. Andrographolide was finally narrowed down as it showed high Hepato-protectant activity (Pa=0.98) and a higher binding energy of -10.43 Kcal/mol (Fig.5).
Molecular dynamics simulation
For a stronger and better insight into the validity of screening strategy, andrographolide-NS5B complex was subjected to molecular dynamic simulation study. The stability of the complex was monitored over the entire simulation time. The lowest binding energy conformation was taken as initial conformation and solvated using SPC (single point charge) water molecule and neutralized by adding chlorine ions. In the first equilibration phase a 30 ps molecular dynamics was performed at 300K and the system was relaxed by 1000 steps of conjugate gradient energy minimization. In the second equilibration phase the system was subjected to 2000 ps dynamics at 300 K and 1 bar pressure. Finally a 5 ns MD simulation was performed to relax the protein-ligand complex. The potential energy fluctuations and RMSD of Cα atoms were monitored. The potential energy of the whole system remained constant throughout the simulation process. The protein structural flexibility during the simulation process was evident from the small drifts observed in the RMSD plot. But the RMSD fluctuations converged after 3 ns simulation proving that the stability was conferred in the protein structure (Fig.6)
A high throughput in silico method curtails the time and cost spent in the synthesis and testing of compounds before entering the clinical trials. Modern drugs have their lineage in traditional medicines. The Indian system of medicines based on herbs is gaining importance globally in recent years because of their efficacy, safety, easy accessibility and cost effectiveness. ADMET has become an integrated part of the drug discovery setup, providing guidance in lead selection and optimization. In our study andrographolide, columbin, esculetin and tinosporide have exhibited the characteristics of a lead compound. In addition to the lead like characteristics, andrographolide has shown good bioactive spectrum as hepatoprotectant. On further enhancement these leads can evolve as promising therapeutic agents against HCV induced hepatic disorders.
Conflict of Interest
We have no conflicts of interest to disclose.
We acknowledge VIT University, India for providing computational facility and support throughout the work.
- Banerjee, Ray RB, Ray R. Onogenic potential of hepatitis C virus proteins. Viruses 2010; 2:2108- 2133.
- Murkowski Z, Heim MH. Interferon signalling in the liver during hepatitis C virus infection. Cytokine 2009; 59:460–466.
- Keyvani H, Fazlalipour M, Monavari SH, Mollaie HR. Hepatitis C virus - proteins, diagnosis, treatment and new approaches for vaccine development. Asian Pac J Cancer Prev 2012; 13:5931-5949.
- Zein NN. Clinical Significance of Hepatitis C Virus Genotypes. Clin Microbiol Rev 2009; 13:223– 235.
- Joyce MA, Tyrrell DL.The cell biology of hepatitis C virus. Microbes and Infect 2010; 12:263-271.
- Mohanapriya A, Sajitha L, Sweta kumari D, Nancy veena kumari. Computational screening and evaluation of bioactive compounds against NS3 helicase of HCV. Int J Pharm Pharm Sci 2013; 5: 370-376.
- Koehn FE, Carter GT. The evolving role of natural products in drug discovery. Nat Rev Drug Discov 2005; 4:206-220.
- Ago H, Adachi T, Yoshida A, Yamamto M, Habuka N, Yatsunami H, Miyano M Crystal structure of the RNA-dependent RNA polymerase of hepatitis C virus. Structure 1999; 7:1417-1426.
- Chen KX, Srikanth Venkatraman, Gopinadhan NA, Zeng Q, Lesburg CA et al. Discovery of SCH 900188: A potent Hepatitis C Virus NS5B polymerase inhibitor prodrug as a development candidate. J Med Chem 2013; 5:244-248.
- Chaudhary GD, Kamboj P, Singh I, Kalia AN. Herb as liver savers – A review. Indian J Nat Prod Res 2010; 1:397-408.
- Govind P. Medicinal plants against liver diseases. IRJP 2011; 2:115-121.
- Handa SS, Sharma A. Hepatoprotective activity of andrographolide from Andrographuis paniculata against carbon tetrachloride. Indian J Med Res 1990; 92:276-283.
- . Doorika P, Ananthi T. Antioxidant and Hepatoprotective properties of Terminalia arjuna bark on Isoniazid Induced Toxicity in Albino rats. Asian J Pharm Tech 2012; 2:15-18
- . Tiwari BK, Khosa RL. Evaluation of the Hepatoprotective and antioxidant effect of Berberis asiatica against experimentally induced liver injury in rats. Int J Pharm Pharm Sci 2010; 2:92-97.
- Vipin Kumar, Modi PK, Saxena KK. Exploration of hepatoprotective activity of aqueous extract of Tinospora cordifolia - an experimental study. Asian J Pharm Clin Res 2013; 6:87-91.
- Patwardhan B, Manish Gautam. Botanical immunodrugs: scope and opportunities. Drug Discov Today 2005; 10:495-502.
- Murthy VN, Praveen Reddy B, Venkateshwarlu V, Kokate CK. Antihepatotoxic activity of Eclipta alba, Tephrosia purpurea and Boerhaavia diffusa. Anc Sci Life 1992;11:182-186.
- Atmaca M, Bilgin HM, Obay BD, Diken H et al. The hepatoprotective effect of coumarin and coumarin derivates on carbon tetrachlorideinduced hepatic injury by antioxidative activities in rats. J Physiol Biochem 2011; 67:569-576.
- Ekbote, Ramesh MT, Mahmood CK, Thippeswamy R et al. Hapatoprotective and antioxidant effect of Azima tetracantha Lam, leaves extracts against CCL4 induced liver injury in rats. Indian J Nat Prod Res 2010; 1:493-499.
- Samuel AJ, Mohan S, Chellappan DK, Kalusalingam A, Ariamuthu S. Hibiscus vitifolius (Linn.) root extracts shows potent protective action against anti-tubercular drug induced hepatotoxicity. J Ethnopharmacol 2012; 141:396–402.
- Bayramoglu G, Bayramoglu A, Engur S, Senturk H et al. The hepatoprotective effects of Hypericum perforatum L. on hepatic ischemia/reperfusion injury in rats. Cytotechnology 2014 ;66:443-448.
- Pramyothin P, Ngamtin C, Poungshompoo S, Chaichantipyuth C. Hepatoprotective activity of Phyllanthus amarus Schum. et. Thonn. Extract in ethanol treated rats: in vitro and in vivo studies. J Ethnopharmacol 2007; 114:169–173.
- Adesokan AA, Akanji MA. Effect of administration of aqueous extract of Enantia chlorantha on the activities of some enzymes in the small intestine of rats. Nigerian J Biochem Molec Biol 2003; 18:103-105.
- Jeyakumar R, Rajesh R, Rajaprabhu D, Ganesan B et al. Hepatoprotective effect of Picrorrhiza kurroa on antioxidant defense system in antitubercular drugs induced hepatotoxicity in rats. Afr J Biotechnol 2009; 8:1314-1315.
- Gao H, Zhou Y. Inhibitory effect of picroside II on hepatocyte apoptosis. Acta Pharm Sin B 2005;26:729–736.
- Saller R, Meier R, Brignoli R. The use of silymarin in the treatment of liver diseases. Drugs 2001; 61: 2035-2063.
- Mondal A, Maity TK, Pal D, Sannigrahi S et al. Isolation and in vivo hepatoprotective activity of Melothria heterophylla (Lour.) Cogn. against chemically induced liver injuries in rats. Asian Pac J Trop Med 2011; 4:619–623.
- . Jagmohan S, Pramod Singh, Rawat B. Chemical constituents and Biological importance of Swertia: A review. Current Research in Chemistry 2011; 3:1-15;
- Vijayan P, Raghu C, Ashok G, Dhanaraj A et al. Antiviral activity of medicinal plants of Nilgiris. Indian J Med Res 2004; 120:24-29.
- Bolton E, Wang Y, Thiessen PA, Bryant SH. PubChem: Integrated Platform of Small Molecules and Biological Activities, in Annual Reports in Computational Chemistry, American Chemical Society, Washington, DC. 2008.
- 1. Sadowski J, Gasteiger J, Klebe G. Comparison of Automatic Three-Dimensional Model Builders Using 639 X-Ray Structures. J Chem Inf Comput Science 1994; 34:1000-1008.
- . Simmonds P, Bukh J, Combet C, Deleage G et al. Consensus Proposals for a Unified System of Nomenclature of Hepatitis C Virus Genotypes. Hepatology 2005; 42:962-973.
- Morris GM, Huey R, Lindstrom W, Sanner MF et al. Autodock4 and AutoDockTools4: automated docking with selective receptor flexibility. J Comput Chem 2009; 30:2785-2791.
- 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;46:3–26.
- Edward HK, Li Di. Drug-like properties: Concepts, Structure Design and Methods: from ADME to Toxicity optimization. Academic Press 2008.
- Ertl P, Rohde B, Selzer P. Fast calculation of molecular polar surface area as a sum of fragment-based contributions and its application to the prediction of drug transport properties. J Med Chem 2000;43:3714-3717.
- . Filimonov D, Poroikov V. Bioactive compound design: Possibilities for industrial use. PASS: Computerized prediction of biological activity spectra for chemical substances, BIOS Scientific Publishers. 1996; 44-56.
- Berendsen HJC, Van der Spoel D, Van Drunen R. GROMACS: A message-passing parallel molecular dynamics implementation. Comput Phys Commun 1995; 91:43-56.
- Schuler LD, Daura X, Van Gunsteren WF. An improved GROMOS96 force field for aliphatic hydrocarbons in the condensed phase. J Comput Chem 2001; 22:1205–1218.
- Schüttelkopf AW, Van Aalten DMF. PRODRG: a tool for high-throughput crystallography of protein-ligand complexes. Acta Crystallogr 2004; D60:1355–1363.
- . Gasteiger E, Hoogland C, Gattiker A, Duvaud S et al. In John M. Walker (ed) Protein Identification and Analysis Tools on the ExPASy Server; 2005.
- Ranjith-Kumar CT ,Cheng Kao C In Seng-Lai Tan (ed) Hepatitis C Viruses: Genomes and Molecular Biology; 2006.
- Waheed Y, Saeed U, Anjum S, Afzal MS et al. Development of Global Consensus Sequence and Analysis of Highly Conserved Domains of HCV NS5B Protein. Hepat Mon 2012; 12.
- . Sarwar MT, Kausar H, Ijaz B, Ahmad W et al. NS4A protein as a marker of HCV history suggests that different HCV genotypes originally evolved from genotype 1b. Virol J 2011; 8:317- 320.
- Dougherty DA. Cation-п Interactions Involving Aromatic Amino Acids. J Nutr 2007; 137:1504S– 1508S.
- Mannhold R, Poda GI, Ostermann C, Tetko IV. Calculation of Molecular Lipophilicity: State-ofthe-Art and Comparison of Log P Methods on More Than 96,000 Compounds. J Pharm Sci 2008; 98:861-893.
- Pajouhesh H, Lenz GR. Medicinal Chemical Properties of Successful Central Nervous System Drugs. NeuroRx® 2005; 2 : 541-553.
- Wager TT, Ramalakshmi Y, Chandrasekaran, Hou X et al. Defining Desirable Central Nervous System Drug Space through the Alignment of Molecular Properties, in Vitro ADME, and Safety Attributes. ACS Chem Neurosci 2010; 1: 420– 434.
- Ertl P, Rohde B, Selzer P.Fast Calculation of Molecular Polar Surface Area as a Sum of Fragment-Based Contributions and Its Application to the Prediction of Drug Transport Properties. J Med Chem 2000; 43: 3714–3717.
- Ghose AK , Viswanadhan VN, Wendoloski JJ. A knowledge-based approach in designing combinatorial or medicinal chemistry libraries for drug discovery. A qualitative and quantitative characterization of known drug databases. J Comb Chem 1999; 1:55-68.
- Ritchie TJ, Macdonald S. The impact of aromatic ring count on compound developability – are too many aromatic rings a liability in drug design? Drug Discov Today 2009; 14:1011-1020.
- Foye WO. Principles of Medicinal Chemistry. Lea and Febiger, Philadelphia. 1998.