Computational Drug Repurposing Resources and Approaches for Discovering Novel Antifungal Drugs against Candida albicans N-Myristoyl Transferase

Candida albicans is a yeast that is an opportunistic fungal pathogen and also identified as ubiquitous polymorphic species that is mainly linked with major fungal infections in humans, particularly in the immunocompromised patients including transplant recipients, chemotherapy patients, HIV-infected patients as well as in low-birth-weight infants. Systemic Candida infections have a high mortality rate of around 29 to 76%. For reducing its infection, limited drugs are existing such as caspofungin, fluconazole, terbinafine, and amphotericin B, etc. which contain unlikable side effects and also toxic. This review intends to utilize advanced bioinformatics technologies such as Molecular docking, Scaffold hopping, Virtual screening, Pharmacophore modeling, Molecular dynamics (MD) simulation for the development of potentially new drug candidates with a drug-repurpose approach against Candida albicans within a limited time frame and also cost reductive.


INTRODUCTION
Invasive infections caused by fungal pathogens are life threatening opportunistic infections, having a high rate of mortality and morbidity in patients. It infects billions of individuals and is responsible for 1.5-2 million deaths annually [1][2][3][4] . Fungal infections have risen dramatically in patients over the last decades which are immunocompromised, because of cancer chemotherapy, solid and hematologic organ transplantation, broad use of antibiotics, surgery, and long-term use of corticosteroids 5 . Invasive fungal infections are particularly exposed to patients receiving cancer treatment, transplant recipients, intensive care unit (ICU) care, and also with acquired immune deficiency syndrome (AIDS). The immunocompromised hosts are having a great risk of these infections with mortality rates from 20% to 40% and it continues to be high, which is relied on what kind of infecting fungal species and the clinical treatment. Several fungal species are present in the world; however, some species, including Candida, Cryptococcus, as well as Aspergillus, lead to life threatening infection in more than 90% of the population. One species of fungus well known as Candida albicans, an ascomycete, and a polymorphic fungus. It is capable of reversibly transforming to various morphologies, include (1) yeast forms, (2) pseudohyphae forms, and (3) true hyphae forms. It's both commensal as well as opportunistic pathogen among humans and ranks as the fourth most common threat of nosocomial bloodstream infections in modern hospitals with roughly 40% death rates 3, 6-17 . Pathogenicity of invasive infection caused by Candida albicans is regulated by several factors namely invasive (a) filamentation, (b) biofilm development, and (c) the ability to escape from the immune system 18 . Studies of metabolic labeling state that Candida albicans synthesize protein N-myristoyl (20-kDa). Myristoyl-CoA: N-myristoyl transferase (NMT), was reported as a target for antifungal as well as antiviral treatment 19 . Antifungal drugs may be used to handle such infections; however, the mortality rates remain high around 50% and there was also a high prevalence of Invasive fungal infections. Discussing treatment options, Antifungals Azoles, echinocandins, and polyenes are existing for the curing of fungal infections which are limited and clinically available 18 . Azoles and polyenes target different biological fungal processes relevant to ergosterol metabolism as well as echinocandins targets cell wall β-1,3 glucan production. 5-flurocytosine is usually used as adjunctive therapy. Fazly et al. described filastatin (a small molecule) that prevents filamentation, adhesion, and virulence of Candida albicans 20 . Garcia et al. reported (N1-(3,5-dichlorophenyl)-5-chloro-2 hydroxybenzamide) halogenated salicylanilide and its analogs Niclosamide, an antifilament molecules that inhibited Candida albicans' biofilm development and had similar antibiofilm and antifilamentation activities 21 . Siwek et al. investigated the antifungal effect of 4-arylthiosemicarbazides and found the isoquinoline-thiosemicarbazide compound to exhibit greater affinity compared to the native ligand 22 . These antifungal agents have significant clinical failures such as unfavorable pharmacokinetic profiles, restricted antifungal range, significant side effects, minimal clinical effectiveness, drug-drug interactions, as well as increased drug-resistance. Therefore it is an urgent need to use all the advanced Bioinformatics tools and techniques to improve the existing fungal drugs or designing novel drugs against it. Existing drugs and the same structural analogous shows the resistant problem on the antifungal targets. Therefore, searching out the new inhibitor is the most promising approach to tackle the resistant fungal infections [23][24][25][26][27][28][29][30][31][32][33][34][35] This review is the effort to use advanced bioinformatics techniques such as Molecular docking, Scaffold hopping, Virtual screening, Pharmacophore modeling, Molecular simulation for developing novel drug candidates with drug repurposing approach against Candida albicans within a short period, cost-reducing and solve the resistant problem in fungal infections.

Candida Albicans: Biofilm Formation
Earlier, microbiologists have studied planktonic cells which are free-floating cells in pure culture. Later they have discovered that there is a link available between sessile cells, microbial pathogenesis, and infections associated with humans and it differs basically from a planktonic cell present in the same species 36 . A broad variety of fungi alternately connecting planktonic cells (freely suspended cells) and multicellular populations, known as biofilms 37 . Biofilms are characterized as well-structured microorganism populations that are interconnected with the surface as well as enclosed by an extracellular matrix (ECM) produced by themselves 38 . The biofilms-associated microorganism is related with several human diseases such as cystic fibrosis, native valve endocarditis and to colonize an extensive range of medical devices which taking into consideration that these structures are very much associated with antimicrobial-resistant and it is very difficult to manage such kind of infections within the clinical setting 39 . A short time ago it has been understood that fungal species form biofilms and it is associated with the escalating clinical problem 38,40,41 . So many Candida Species have been identified but the most famous studied species is Candida albicans as a well-developed biofilm activity with the most adaptable opportunistic pathogen 42 . There have been so many Candida species reported, but as a well-developed biofilm operation with the most adaptable opportunistic pathogen, the most popular species studied is Candida albicans. Biofilm is developed on various medical devices such as dentures, neurosurgical shunts, speech prostheses, breast implants, prosthetic joints, endotracheal tubes, intracardiac prosthetic devices, urinary catheters, dialysis catheters for peritoneal and hemodialysis, peripheral and venous catheters 43 . It exists in various types, such as yeast, hyphae, and multicellular biofilm 44 . Candida albicans adherence and colonization to denture acrylic substrates as well as oral mucosa is the first step of pathogenesis [45][46][47][48] . Candida albicans' initial attachment to the surface is limited by the pH, osmolarity, flow of the nearby medium, such as urine, antimicrobial agents, bacteria, saliva, Mucus, temperature, blood, as well as the host immune factors 49-54 . Candida albicans biofilm formation having different phases of development. It contains substrate adhesion, colonization, extracellular material production, and maturation 53-62 . Biofilm development has been shown in Fig. 1.
The yeast cell's ability to shape biofilms on the implanted medical devices or on the surface in the host enhances its virulence. Candida albicans adhere to the surface with the support of Eap1p (cell wall protein) and Als3p (agglutinin like sequence protein) 63,64 . Als3p and Eap1p are initiations to the formation of microcolonies and further Efg1 regulatory protein is essential for the production of biofilm and its development of pseudo-and true-hyphae to form a complex association of hyphal structures with budding yeast-like cells spread throughout 52,65 . Further, the growth and the maturation of Biofilms, Candida albicans biofilm cells encompass a betaglucan rich extracellular matrix that protects from environmental stresses, antimicrobial agents, and host defenses 66 . The existence of the hypoxic environment is correlated with the maturation of biofilm and this condition induces Tye7p-dependent up-regulation of glycolytic genes required to respond to hypoxia and prevent uncontrolled hyphal formation 67 . In the final step, planktonic yeast cells dispersed from the mature biofilm and established a new colony on a new surface to grow a new biofilm from Candida 68 . Candida albicans Biofilm formation has been presented in Fig. 2.
The diverse transcription factors such as Efg1p, Ace2p, Zap1p, and Bcr1p are the regulator which controlled the formation of Biofilm 63, 69-71 . The various genes have been presented in Fig. 2, for controlling and maintaining the development of biofilm. The most important thing is to understand the mechanism of those genes so inhibition such kind of infections in the populations. The key sites of the infections are biomaterials 43   In the right-hand side part of the diagram, the genes are connected and involve in pathway but in the left hand side part, the genes may not attach to an established pathway but function in a particular step. Arrows signify positive connection, the Dashed line signify repression by an indirect mechanism. "+" sign indicates that an upstream gene stimulates the expression of the downstream target and "-" sign is opposite of it. "T-shaped" indicated a negative relationship (repression by an indirect process).

Candida albicans: NMT [N-Myristoyl-transferase] as a drug target
Post-translational modification is a very important step for proteins to function in a specific way through further modification. Post-translational modification occurs at the protein's C-or N-terminal or on the amino acid side chain 80 . The modification occurs during the post-translational modification are different according to the different transformation such as C terminal amidation, N-terminal acetylation, phosphorylation of threonine, tyrosine or serine residues mainly in kinases, methylation of arginine and lysine residues mostly in histones, acylation of lysine residues and oxidation mainly in proline residues 81 . A less common type of posttranslational modification is lipidation. Lipidation is the covalent attachment of the lipid moiety to the protein. Lipidation increases stability, membrane interaction, protein hydrophobicity, changes in conformation, trafficking, etc. Different types of lipidation are known, differing according to the group being attached and the position of attachment 82 . Lipidation attachment has been presented in (Table 1).

Researchers did extensive research and identified
Candida albicans as an antiviral and antifungal therapy target. N-myristoyl transferase is indeed a monomeric cytosolic enzyme that is vital for the function and growth of fungi 88, 89 . NMT is present in eukaryotes such as animals, protozoa, and fungi excluding bacteria. Protein N-myristoyl transferase is associate with the Gcn5-related N-acetyltransferases superfamily 90 . Candida albicans NMT contains 451 residues of amino acids and 45% of the human enzyme sequence identity. NMT is a compact globular, wedge-formed structure in which a big saddle-shaped betasheet present and it occupies the center of the protein structure, also, it is surrounded by several helices means consisting of an N-terminal strand, preceded by two helices, three anti-parallel beta strands, preceded by a signature (central helix) and last beta-strand 91 . The NMT protein structure has been illustrated in Fig. 3. C-terminal half is crucial for the peptide binding site and N-terminal half is important to form mainly Myristoyl-CoA binding site 90 . N-Myristoyl-transferase catalysis reaction is catalyzed by N-myristoyl-transferase, the co-translational addition of myristic acid (14-C saturated fatty acid) to the N-terminal Glycine (GLY) residue of the substrate protein via amide bonding. The N-myristoyl transferase catalysis reaction is performed by the ordered Bi-Bi reaction mechanism, the enzyme forming a high binary selectivity complex (Myristol-CoA-NMT). This binary complex is essential to the further interaction of N-Myristoyltransferase with peptide and produces a ternary complex recognized as NMT-Myristoyl-CoA-Peptide, following the catalytic transfer of myristate to the peptide substrate. The first free CoA is released, followed by the N-myristylated protein.
In general, Myristoylation is irreversible as well as a significant post-translational modification is defined as N-terminal lipidation of eukaryotic and viral proteins 92 . Myristoylation mechanism has been shown in Fig. 4.
Myristoylation involved in anchoring and directing proteins to membranes and their effects such as signal transduction, cellular regulation, numerous pathologic processes caused by viruses, apoptosis, and translocation 93,94 . The binding of  myristoyl residues enables hydrophobicity to affect protein partitioning to the cell membrane and promote the interaction of protein-proteins. It is important for the overall biological expression of viral and cellular protein activity 95-97 . In Fungi, the myristoyl is associated with the cellular membrane and myristoyl-protein interactions. This protein takes part in protein and vesicular trafficking and signals transduction cascade. In Candida albicans with defective NMT unable to infect mice 98, 99 . Genetic studies showed that the enzyme is important for pathogenic Candida albicans to grow vegetatively 100 . NMT is a good antifungal agent target because it is responsible for systemic fungal infections and lacking its expression is related to significantly decrease cell growth and increased cell death.

Candida albicans: NMt inhibitors
Several potent and selective inhibitors have been identified against Candida albicans NMT which showed low inhibitory activity against hNMT. All NMT polypeptides have similar folding but different inhibitor binding sites because of their particular amino acid differences 90,101-104 . As we studied earlier that NMT is responsible for the survival and growth of diverse fungal species, therefore so many different inhibitors have also been identified for reducing its fungal activity such as Benzofurans inhibitors 101-103 , Benzothiazole inhibitors 104 , Myristic acid analogs 105,106 , Peptidomimetic inhibitors 88,107 , p-toluene sulphonamide inhibitors 108 , etc. Devadas et al. reported a peptidomimetic inhibitor against Candida albicans NMT. This inhibitor was structured dependent on octapeptide substrate GLYASKLS-NH2 that was obtained from the N-terminal fragment of ARF2 (ADP ribosylation factor 2) and its analogous ALYASKLS-NH2 107    in vitro but not showed in vivo 106 . A new class of inhibitor was also reported named as p-toluene sulfonamides 110 .
In the journey of finding out the novel compounds with high selectivity, Benzofurans and Benzothiazoles are more promising than the previously reported compounds [102][103][104] . From the viewpoint of the development of antifungal drug candidates, other inhibitors were also developed such as 4-arylthiosemicarbazides derivatives 22 , novel benzofuran-semicarbazide hybrids and 1,3-dialkoxybenzene-semicarbazide hybrids, etc. 111 . Candida albicans NMT inhibitors are presented in Fig. 5.

MATERIALS AND METHODS
Computational drug discovery approaches really works for finding out the novel agents as new medications. Day by Day pharmaceutical and biotech companies are growing to help the society but the major problems we have to face today is the cost of the drugs are increasing and the expenses which we have to pay for the medicine are increasing. The drug productivity measures are unable to meet the increasing demands. Thus, advanced bioinformatics tools and techniques were discussed to find out the novel antifungal agents using drug repurposing approach.

Virtual Screening and molecular docking
Protein-ligand docking is a technique commonly used to determine a drug candidate's binding orientation to their specific target. In our survey, we are in consideration of Candida albicans NMT as a drug target for drug designing purpose. Typically molecular docking technique is performed either to searched out that how a specific ligand molecule bind to a target protein or illustrate binding interaction with the targetspecific amino acid residues either H-bonding, Hydrophobic interaction, disulfide bond formation, salt bridge, pi-pi interaction or to find out the potent compound from the available databases that can bind with the target protein [112][113][114][115][116][117][118][119] . The docking can be categorized into two key steps, the initial positioning of the ligand structure at the active site of the target protein using the docking algorithm. followed by uses of the scoring function to assess the potency of the binding interaction. There are a huge number of docking algorithms, tools, techniques are available to highlight the diverse orientation of the interaction between the ligand and the target structure as shown in (Table 2).
In the early days, the docking algorithm did not treat the protein and ligand as flexible objects, only the six translations and also the rotational degree of freedom was incorporated. Currently, we are having more consistent docking methods which give the options for flexible docking that the target protein is treated as fixed during the docking process, but the ligand is capable to move around the target.
In this situation, the active site of the protein shall not be considered to undergo any significant changes in conformity with the binding of the ligand. The flexible docking is broadly used with parallel computing resources to relatively, accurately, and quickly search databases for potential ligands to a target protein. The more precise algorithms which consider both ligand and receptor flexibility are very time-consuming, therefore have not been developed extensively. The algorithm which treats flexibility of the ligand is partitioned into three categories, for example, stochastic methods, systematic methods, and simulation methods. The docked poses are ranked and assessed using docking scoring functions which estimate a ligand's binding free energy to a receptor, which is very important to differentiate the right poses from incorrect ones. The scoring function incorporates diverse sorts of terms that express electrostatic interactions, solvation effects, non bonded interactions, and van der Waals interactions 128 . The structure-based virtual screening framework was presented in Fig. 6.

Pharmacophore modeling
The initial idea of a pharmacophore was developed by Paul Ehrlich during the late 1800s. The theory in the past was that in a molecule there were some chemical groups or functions that were responsible for a biological effect and that certain effect molecules even had similar functions. Fig. 7, revealed the pharmacophore, with its applications. Later in 1960, Schueler coined the term pharmacophore in his book "Chemobiodynamics and Drug Design". It elucidated that a molecular structure that expresses the essential characteristics liable for the biological activity of the drug. The Pharmacophore has been illustrated by IUPAC since 1997 129 .    It projected that a pharmacophore is the collection of features known steric and electronic that are original to ensuring the mainly desirable supramolecular contacts linked with the desired protein target and blocking its biological activity 130 . The pharmacophore reveals an abstract idea. it relies on the features shared by a group of active molecules or it is the pattern of the features of a molecule that is responsible for a biological effect. Forms of molecular features patterns are hydrogen-bond acceptors, hydrogen-bond donors, hydrophobic, anionic, cationic, aromatic plus any such type of possible combinations presented in Fig. 8.

Pharmacophore modeling (Structure-based)
The structure-based approach to pharmacophore modeling describes the relevant presentation of important interactions in a proteinbinding pocket. This pharmacophore modeling is appropriate in aspects of a free structure or a complex target-ligand structure. The free structure is classified as apo, and the holo known as the target-ligand complex. The structural pharmacophore modeling was performed using free ligand without protein, using only proteinactive site details and when the pharmacophore modeling uses protein-ligand structure complexes utilize the possible interactions involving protein and ligand, shown in Fig. 9. Structure-based pharmacophore modeling is a very effective tool for virtual screening such as multi-target drug design, scaffold hopping, parallel screening, QSAR, and multi-target drug development 131,132 .

Pharmacophore modeling (Ligand-Based)
Pharmacophore modeling based on ligand structure is a powerful computational tool of great importance for helping to discover a new drug compound. It is done by extracting the important and crucial chemical features, among the set of ligands. The ligands have been divided into training and test for alignment and generating a pharmacophore model, presented in Fig. 10. This model can be utilized further for the virtual screening process for finding a similar feature molecule that behaves like a drug 133, 134 .

Software available for performing pharmacophore modeling
There are diverse software and tools are available to perform structure and ligandbased pharmacophore modeling, presented in (Table 3-4). Scaffold Hopping Schneider et al. (1999) presented Scaffold hopping, in 1999. It is a method for the discovery   154,155 . It is one of those approaches for finding out the new lead candidates 156 . Scaffold hopping intends to discover a structurally novel substance structure starting from previously identified active compounds by altering the center core structure of the molecule 157 . Scaffold hopping is commonly used in lead optimization. Since using HTS, so many compounds are unsuccessful compounds with poor PK properties and poor physiochemical properties. To overcome this, side-chain modification is sufficient sometimes, the core structure of the parent molecule or the scaffold may often be changed 152,[158][159][160] . Why Scaffold hopping is so important • Central scaffolds are also specifically involved in target protein interactions. An enhanced binding affinity can result from a change in the scaffold. • Replacing a lipophilic scaffold by an extra polar one could enhance the solubility of the compounds (lipophilic compound soluble in fats, oils, lipids etc. for increasing solubility, lipophilic scaffold can be change by an extra polar side chain or fragment). • Replacing a very flexible scaffold known as peptide backbone by an inflexible central scaffold would also considerably advance the binding affinity and on the total DMPK characteristics. • Changes in the core of the structure may lead to a patentable novel compound. • Replacing a metabolically labile scaffold via a reduced amount of toxic, and an additional stable one will improve pharmacokinetic properties.

Insilico approaches for scaffold hopping
There are different approaches are available for scaffold hopping but the main idea behind is (1) matching of shape, (2) searching for pharmacophores, (3) replacement of fragments, (4) looking for similarities, and (5) machine learning, etc. The Shape matching approach describes if the compounds are structurally related means display similar biological activities and if the compounds are more distantly related, the less probable to show the same biological effects. In shape matching, it is possible to find out the compounds which mimic accurately this structure, but it is not possible to get the same features that are significant for binding to the target structure 161,162 . If the ligands are structurally different however, can adopt similar shape and share common features, It is possible to derive 3D pharmacophores, which is used for shape matching scaffold hopping.
But the drawback is for searching out the 3D pharmacophore using different chemical structure databases is not sufficient to find out the novel scaffold because it searches from the known compound databases [163][164][165][166] . Another approach to scaffold hopping is fragment replacement, in which no need for the replacement of the entire compounds but searching for a replacement of fragment of an active compound [167][168][169] . Similarity searching is also used for scaffold hopping. The chemical structures are assembled in these algorithms using fragment joining as well as the novel scaffolds are resolved by their match to the query [170][171][172] . Also, machine learning methods are used for scaffolds hopping, methods together with self-organizing maps that allow compound distributions to be visualized 173,174 . The computational methods of scaffold hopping are shown in Fig. 11. The software tools is listed out in (Table 5).

Molecular Simulations And Advancement
Since the first protein (folded globular protein) MD simulation is discussed in 1977 186 . In December 1999, IBM declared a five-year intend to build up a massively parallel computer for studying biomolecular phenomena, in which they Core Hopping The core-hopping technique is to test several possible scaffolds 177 as protocores) against a template and look for alignments of possible (also known attachment points on the scaffold with the attachment points on the template.

4.
LigCSRre LigCSRre is a modern effective and standardized method for 3D 178 matching screening of tiny compounds, the modular plan of which opens the door to lots of improvements.

e-LEA3D
The approach is perfectly appropriate for scaffold-hopping, this section 179 moreover permits a search for potential substitutes to a selected scaffold. 6.
ChemMapper ChemMapper using the user given the chemical structure of the molecules 180 as the query, the highest alike structure in respect of 3D similarity is sent back using related pharmacology annotations. 7.
SHOP It is a grid-based technique for Scaffold bouncing. In a database, scaffolds 181 were predictable utilizing 3 types of 3D-descriptors. 8. LeadGrow+ Creating a molecular library for efficient scaffold hopping. 182 9.
Recore Recore is a rapid and flexible scaffold hopping method based on 183 conformations of small molecule crystal structures. 10.
HTSFPs (HTSFPs), It is a method that matches patterns of actions in investigational 184 screens.

11.
MORPH MORPH is a software tool for scaffold hopping which can scientifically 185 alter aromatic rings in molecular 3-dimensional models exclusive change of the non-hydrogen atom co-ordinates in the rings.
have discussed for increasing longer simulation time as well as developing computing software and hardware for bimolecular MD simulation 187 . Certain software packages were also developed for simultaneously scales well-organized MD simulation very well on machines 188 . A massively parallel machine such as Anton was introduced which was able to reach millisecond time-scale simulation for biomolecular systems 189 . In recent times graphical processing unit which is known as GPU achieved remarkable progress with high-performance computing capability for MD simulations 190,191 . currently, MD simulation is very important for studying protein, DNA, and RNA systems. In the MD simulation so many terms are used in which force field is very significant where a protein force field included bonded (bond angle, dihedral angles, bond length) and also non-bonded interactions (electrostatic, van der Waals). The development of improved sampling methods and escalating computational performance came with more inaccuracies in the protein force field 192 . In these aspects, the classical protein field has been improved with Gromacs 188 ,AMBER 193 ,CHARMM 194 ,and NAMD 195 . A basic algorithm for MD simulation has been represented Fig. 12 and a list of software has been shown in (Table 6). apart from this MD simulation software, there are other software also available such as Desmond 196 ,TINKER,DL_POLY 197 ,ESPResSo 198 , etc. so that it is understood that MD is already an important tool in serving to understand biology.

RESULTS AND DISCUSSION
The anticipated outcome could be development and searching out the novel, specific inhibitors for Candida albicans MyristoylCoA: Protein N-Myristoyltransferase as anti-fungal agents using advanced computational approaches. A humble beginning made towards this end needs patronage for further development. All the more interesting on this aspect is, still as on date, no successful attempt has been made towards development of a best, specific inhibitor for Candida albicans MyristoylCoA: Protein N-Myristoyltransferase which again augments support.

CONCLUSION AND FUTURE PERSPECTIVE
N-myristoyl transferase is a monomeric cytosolic protein that is vital for the function and growth of fungi. There are so many inhibitors that have been designed against Candida albicans NMT for reducing fungal infections in humans but at present antifungal drugs are not perfect in the expressions of the antifungal spectrum, efficacy, and protection. Drug repurposing is one of the most significant, more affordable, and increasingly proficient techniques in drug discovery. So right now, we have examined in silico drug repurposing approach which joins Molecular docking, Virtual Screening, Pharmacophore demonstrating, Scaffold hopping, and Molecular dynamics (MD) simulation for the advancement of a novel Candida albicans NMT inhibitors.

ACKNOWLEDGMENTS
The Authors would like to offer their sincere thanks to ICMR (Indian Council of Medical Research), New Delhi, Government of India, for financial support related help For the present research and Department of Biotechnology (DBT) New Delhi for offering help as Bioinformatics infrastructure facility. The authors also would like to acknowledge the Schrödinger group for giving software support.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.

AUTHORS' CONTRIBUTION
All authors have made substantial, direct,

DATA AVAILABILITY
All datasets generated or analyzed during this study are included in this manuscript.

ETHICS STATEMENT
Not applicable. Siwek A, Stefanska J, Dzitko K, Ruszczak A. Antifungal effect of 4-arylthiosemicarbazides against Candida species. Search for molecular basis of antifungal activity of thiosemicarbazide derivatives. J Mol Model.