No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Development and Validation of AMANDA, a New Algorithm for Selecting Highly Relevant Regions in Molecular Interaction Fields
Abstract
Descriptors based on Molecular Interaction Fields (MIF) are highly suitable for drug discovery, but their size (thousands of variables) often limits their application in practice. Here we describe a simple and fast computational method that extracts from a MIF a handful of highly informative points (hot spots) which summarize the most relevant information. The method was specifically developed for drug discovery, is fast, and does not require human supervision, being suitable for its application on very large series of compounds. The quality of the results has been tested by running the method on the ligand structure of a large number of ligand-receptor complexes and then comparing the position of the selected hot spots with actual atoms of the receptor. As an additional test, the hot spots obtained with the novel method were used to obtain GRIND-like molecular descriptors which were compared with the original GRIND. In both cases the results show that the novel method is highly suitable for describing ligand-receptor interactions and compares favorably with other state-of-the-art methods.
... The LB SHOP method was validated on a number of X-ray complexes chosen from three protein families (thrombin, HIV protease, and influenza virus neuraminidase). For each family, complexes were chosen such that the ligands had comparable binding mode and interactions with the target, and that scaffold replacement could be carried out using a variable number of anchor points (2)(3)(4). 44 A training set database was assembled from 471 scaffolds extracted from previously published combinatorial libraries, manually introducing anchor points as dummy atoms. For each scaffold, 3D conformations were generated. ...
... 68 The only step which requires user intervention is ligand fragmentation into R-groups. 62 The procedure for generating a topomer from a monovalent molecular fragment is briefly outlined in the following: (1) cap the open valence of the monovalent fragment with a suitable group (e.g., a methyl) and generate a single 3D conformation out of the 2D topology, considering all ionizable groups as being in their uncharged form; (2) translate and rotate the model to overlay the bond joining the fragment to the cap onto a vector fixed in Cartesian space; (3) adjust acyclic torsions, invert stereocenters, and standardize ring puckering as required to obtain consistent topomer geometries as determined by a set of rules; and (4) remove the cap. Regarding chirality inversion, while the original stereochemistry is stored within the fragment metadata, the stereochemistry in 3D geometry is actually rule-based and independent from the original one, according to the aforementioned trade-off between consistency and accuracy; even eventual internal steric clashes are ignored. ...
... The ability of AMANDA to extract relevant ligand hotspots was validated with objective metrics on a set of $ 800 complexes taken from the PDBbind database, 47 showing a satisfactory ability to identify the interaction partners in the protein just based on hotspot extraction from ligand MIFs. 2 The subsequent step involves the computation of auto-(within the same MIF) and cross-(across different MIFs) correlograms between hotspot pairs. For each pair the product of MIF energies is computed and binned based on the discretized distance between the nodes (Fig. 20). ...
Molecular interaction fields (MIFs) were introduced in computational and medicinal chemistry in the late 1970s to describe the three-dimensional interactions between a molecule and its environment. Molecules which elicit similar interactions generate similar MIFs, even though their underlying chemical structures may differ quite substantially. This paradigm enables a number of core tasks in medicinal chemistry, ranging from bioisosteric replacement and molecular similarity assessment to virtual screening, quantitative structure–activity relationships, and prediction of metabolic liability. This article provides an overview of the various MIF-based in silico methodologies that have been applied to both ligand- and structure-based molecular designs.
... All compounds were converted to sdf format and then imported. Compounds were imported into the Pentacle QSAR software [36], protonated at pH 7.4, and oriented according to the principal moments of inertia. IC 50 values were converted to pIC 50 and used as activities. ...
... IC 50 values were converted to pIC 50 and used as activities. Standard GRIND descriptors were calculated [36,37], based on molecular interaction field (MIF) probes: DRY (representing hydrophobic interactions), N1 (representing neutral flat NH like in amide, as hydrogen bond donor), O (sp2 carbonyl oxygen, representing hydrogen bond acceptor), TIP (molecular shape descriptor) and the following principal components analysis (PCA) and PLS models were constructed. The number of PC components in the PCA model was five, and in PLS the number of latent variables (LV) was 3. Fractional factorial design (FFD) variables selection was repeated until R 2 and Q 2 values of the PLS model converged. ...
... The EIIP filtered compounds were loaded into the model and their pIC 50 activities were calculated. Calculations were done in Pentacle 1.05 [36][37][38]. ...
Arginase, a drug target for the treatment of leishmaniasis, is involved in the biosynthesis of polyamines. Flavonoids are interesting natural compounds found in many foods and some of them may inhibit this enzyme. The MetIDB database containing 5667 compounds was screened using an EIIP/AQVN filter and 3D QSAR to find the most promising candidate compounds. In addition, these top hits were screened in silico versus human arginase and an anti-target battery consisting of cytochromes P450 2a6, 2c9, 3a4, sulfotransferase, and the pregnane-X-receptor in order to flag their possible interactions with these proteins involved in the metabolism of substances. The resulting compounds may have promise to be further developed for the treatment of leishmaniasis.
... To assess the pharmacological activity of our candidates, similarly as in the recently reported study on COVID-19 drug repurposing, [11] we carried ligand-based screening, calculating the molecular similarity using Principal Component Analysis (PCA) based on Molecular Interaction Fields (MIF) descriptors. [18] All 44 selected drugs from the previous step (ISM-SM) were imported in Pentacle software, [19] protonated at pH 7.4, and aligned towards the principal moment of inertia. In ligandbased virtual screening, we used the centroid distance method as criteria for the similarity between co-crystallized inhibitors GRL0167, XR8-89, VBY501, and candidate compounds. ...
... Applied probes (mapped regions of molecule surface) were DRY (hydrophobic interactions) O (hydrogen bond acceptor) N1 (hydrogen bond donor) and TIP (molecular shape descriptor). The discretization Method was AMANDA [18] with a scale factor of 0.55. The encoding Method was MACC2 and weights were the following: DRY: À 0.5, O: À 2.6, N1: À 4.2, TIP: À 0.75. ...
The need for an effective drug against COVID-19, is, after almost 18 months since the global pandemics outburst, still very high. A very quick and safe approach to counteract COVID-19 is in silico drug repurposing. The SARS-CoV-2 PLpro promotes vi-ral replication and modulates the host immune system, resulting in inhibition of the host antiviral innate immune response, and there-fore is an attractive drug target. In this study, we used a combined in silico virtual screening candidates for SARS-CoV-2 PLpro protease inhibitors. We used the Informational spectrum method applied for Small Molecules for searching the Drugbank database and further followed by molecular docking. After in silico screening of drug space, we identified 44 drugs as potential SARS-CoV-2 PLpro inhibitors that we propose for further experimental testing.
... To assess the pharmacological activity of our candidates, similarly as in the recently reported study on COVID-19 drug repurposing, [11] we carried ligand-based screening, calculating the molecular similarity using Principal Component Analysis (PCA) based on Molecular Interaction Fields (MIF) descriptors. [18] All 44 selected drugs from the previous step (ISM-SM) were imported in Pentacle software, [19] protonated at pH 7.4, and aligned towards the principal moment of inertia. In ligandbased virtual screening, we used the centroid distance method as criteria for the similarity between co-crystallized inhibitors GRL0167, XR8-89, VBY501, and candidate compounds. ...
... Applied probes (mapped regions of molecule surface) were DRY (hydrophobic interactions) O (hydrogen bond acceptor) N1 (hydrogen bond donor) and TIP (molecular shape descriptor). The discretization Method was AMANDA [18] with a scale factor of 0.55. The encoding Method was MACC2 and weights were the following: DRY: À 0.5, O: À 2.6, N1: À 4.2, TIP: À 0.75. ...
In the current pandemic, finding an effective drug to prevent or treat the infection is the highest priority. A rapid and safe approach to counteract COVID‐19 is in silico drug repurposing. The SARS‐CoV‐2 PLpro promotes viral replication and modulates the host immune system, resulting in inhibition of the host antiviral innate immune response, and therefore is an attractive drug target. In this study, we used a combined in silico virtual screening for candidates for SARS‐CoV‐2 PLpro protease inhibitors. We used the Informational spectrum method applied for Small Molecules for searching the Drugbank database followed by molecular docking. After in silico screening of drug space, we identified 44 drugs as potential SARS‐CoV‐2 PLpro inhibitors that we propose for further experimental testing. Combined in silico machinery, consisting of long‐ and short‐range molecular interaction methods is capable of filtering large compound databases, yielding a good candidate for a receptor. The advantage of this approach is a fast assessment of a candidate, starting from basic information on both protein and small organic molecule – FASTA sequence and SMILES notation.
... Minimized structures of compounds were transferred to Pentacle 1.05 software (Molecular Discovery Ltd., Oxford, UK) to generate the GRid-INdependent Descriptors (GRINDs). GRIND descriptors specifically describe the pharmacodynamic properties, including receptor-ligand interactions [47][48][49][50]. Amanda algorithm was applied to extract the alignment-independent descriptors (GRINDs) [48]. ...
... GRIND descriptors specifically describe the pharmacodynamic properties, including receptor-ligand interactions [47][48][49][50]. Amanda algorithm was applied to extract the alignment-independent descriptors (GRINDs) [48]. This algorithm works through the following steps: ...
Dihydrofolate reductase (DHFR) is an essential enzyme that participates in folate metabolism and purine and thymidylate synthesis in cell proliferation. It converts dihydrofolate (DHF) to tetrahydrofolate (THF) in the presence of nicotinamide adenine dinucleotide phosphate (NADPH). This enzyme is found within all organisms adjusting the cellular level of THF. Negligible DHFR activity leads to a deficiency of THF and cell death. This trait is used to inhibit the cancer cells. DHFR has been extensively studied as the therapeutic target for cancer treatment. Accordingly, there is an urgent need for the identification and development of novel effective inhibitors with higher selectivity, lower toxicity, and better potency than currently available drugs. Hence, we aimed to identify new compounds by utilizing the alignment-independent three-dimensional quantitative structure-activity relationship (3D-QSAR) and structure-based pharmacophore modeling. Using the results obtained from these approaches, several antagonists have been retrieved from the virtual screening by applying some filters such as Lipinski’s rule of five. Selected compounds were then docked into the binding site of the receptor for identification of ligand-receptor interactions, binding affinity prediction, and refinement based on GoldScore fitness values. Eventually, pharmacokinetic/drug-likeness features and toxicity profiles of novel compounds were predicted and evaluated. Four hits with PubChem CIDs of 136138676, 94182663, 60219817, and 133300845 were finally proposed as new candidates with potential inhibitory activity against human DHFR.
... 43 GRINDs have been calculated by ALMOND and AMANDA algorithms. 44,45 Both algorithms work through 3 main steps: (i) calculating molecular interaction fields (MIFs) followed by identification of regions with desirable interaction energies, (ii) MIF filtration to focus on the regions with the most favorable interaction energies. This step differs in ALMOND and AMANDA. ...
... The approach provides data that connect directly with the initial molecular structures. 44,45 Herein, probes of DRY (hydrophobic interactions), O (carbonyl oxygen as a hydrogen bond acceptor), the N1 (amide nitrogen as a hydrogen bond donor), and TIP (the shape between the ligand and the protein) have been applied to calculate MIFs. 47 Default parameters were used for the distance between grid points (0.5 Å), the number of extracted nodes (100 for each MIFs), and the smoothing window (0.8 grid unites/0.4 ...
Dihydrofolate reductase (DHFR) is an essential enzyme in the folate metabolism pathway and an important target of antineoplastic, antimicrobial, antiprotozoal, and antiinflammatory drugs. Despite the clinical effectiveness of current antifolate treatments, new drugs are needed to be designed due to developing resistance of this enzyme through multiple‐site mutagenesis. Understanding the factors affecting the ligand binding selectivity profiles among DHFR families is critical for the design of novel potent and selective inhibitors, with the least side effects, against DHFR of pathogens. Hybrid scaffolds containing pyrimidine ring are effective in DHFR inhibition. In this study, using proteochemometric (PCM) modeling, we designed and evaluated new potent pyrimidine scaffold‐based inhibitors via 3‐dimensional alignment‐free GRid‐INdependent Descriptors (GRIND), VolSurf molecular, and sequence‐based (z‐scale) descriptors to provide ligand and receptor descriptors, respectively. Validation and robustness of the model were confirmed by venetian blinds cross‐validation and Y‐scrambling approaches, respectively. Applicability domain (AD) analysis was performed to estimate the likelihood of reliable prediction for compounds. To show the applicability of the PCM model, new ligands were designed using structural data retrieved from this model. Inhibitory activities of the designs were then predicted, and selectivity ratio profiles were investigated. Finally, potent and highly selective inhibitors were identified regarding the protozoan parasite Toxoplasma gondii, followed by evaluating the ADMET parameters of the ligands.
... AMANDA algorithm (Durán et al., 2008) was applied for the extraction of the most relevant MIF using default energy cut-offs −0.5, −2.6, −4.2, −0.75 values for DRY, O, N1, and TIP probes, respectively. Consistently Large Auto And Cross Correlation (CLACC) algorithm was used to encode the prefiltered nodes into four auto (DRY-DRY, O-O, N1-N1, TIP-TIP) and six cross (DRY-O, DRY-N1, DRY-TIP, O-N1, O-TIP, N1-TIP) correlograms (Durán Alcaide, 2010). ...
... Thus, it reflects the predictive ability of structurally diverse data set of hERG blockers. Methodologically, in the present study highly significant GRIND variables have been selected with the help of AMANDA algorithm (Durán et al., 2008) as described by Pastor et al. (Pastor et al., 2000) in comparison with ALMOND algorithm utilized by most of the previous investigations that select a fixed number of variable irrespective of strength of the MIFs. Furthermore, in present investigation more robust pharmacophore models have been developed by normalizing suboptimal transport factor of the data and selecting templates with best pIC 50 /clogP and pIC 50 /molecular weight ratio. ...
The hERG (human ether-a-go-go-related gene) encoded potassium ion (K⁺) channel plays a major role in cardiac repolarization. Drug-induced blockade of hERG has been a major cause of potentially lethal ventricular tachycardia termed Torsades de Pointes (TdPs). Therefore, we presented a pharmacoinformatics strategy using combined ligand and structure based models for the prediction of hERG inhibition potential (IC50) of new chemical entities (NCEs) during early stages of drug design and development. Integrated GRid-INdependent Descriptor (GRIND) models, and lipophilic efficiency (LipE), ligand efficiency (LE) guided template selection for the structure based pharmacophore models have been used for virtual screening and subsequent hERG activity (pIC50) prediction of identified hits. Finally selected two hits were experimentally evaluated for hERG inhibition potential (pIC50) using whole cell patch clamp assay. Overall, our results demonstrate a difference of less than ±1.6 log unit between experimentally determined and predicted hERG inhibition potential (IC50) of the selected hits. This revealed predictive ability and robustness of our models and could help in correctly rank the potency order (lower μM to higher nM range) against hERG.
... Gasteiger-Huckel method was used to calculate partial atomic charges of all compounds. The final structures were subjected to Pentacle 1.05 software (22) and grid independent descriptors (GRINDs) were calculated for them using both ALMOND and AMANDA algorithms (23,24). Both algorithms work through three following steps: (a) calculating MIFs for different types of interactions, (b) node filtration process in which regions with greatest favorable interaction energy are selected. ...
... The main difference between ALMOND and AMANDA relates to the node filtration step. ALMOND uses a Fedorov-like optimization algorithm (25) to reduce the number of nodes, whereas AMANDA uses a pre filtering step in which all nodes failing an energy cutoff are primarily removed (24), and (c) finally, the chosen nodes are encoded into the descriptors. Pairs of interaction energies are multiplied and the greatest product is kept for each internode distance. ...
Isoform diversity, critical physiological roles and involvement in major diseases/disorders such as glaucoma, epilepsy, Alzheimer's disease, obesity, and cancers have made carbonic anhydrase (CA), one of the most interesting case studies in the field of computer aided drug design. Since applying non-selective inhibitors can result in major side effects, there have been considerable efforts so far to achieve selective inhibitors for different isoforms of CA. Using proteochemometrics approach, the chemical interaction space governed by a group of 4-amino-substituted benzenesulfonamides and human CAs has been explored in the present study. Several validation methods have been utilized to assess the validity, robustness and predictivity power of the proposed proteochemometric model. Our model has offered major structural information that can be applied to design new selective inhibitors for distinct isoforms of CA. To prove the applicability of the proposed model, new compounds have been designed based on the offered discriminative structural features.
... In order to shed light on 3D structural features involved in CYP3A4 inhibition and stereoselectivity, the most probable binding conformation of each ligand obtained from molecular docking, along with their biological activities were imported in Pentacle version 1.05 (Durán et al. 2008). Molecular Interaction Fields (MIFs) were calculated by employing four different probes: ...
... Where Elj is Lennard-Jones energy, Eel is electrostatic energy and Ehb is hydrogen bond energies at that point. Furthermore, most important and relevant MIFs were selected by using AMANDA algorithm in pentacle (Durán et al. 2008). Consistently large auto and cross correlation algorithm was used for encoding the final nodes into GRIND thus, producing consistent sets of variables. ...
CYP3A4 has been identified as a major enzyme involved in the metabolism of drugs and xenobiotics in the human body. It is mainly expressed in the liver and has a wide capacity to oxidize structurally and functionally diverse compounds due to its large promiscuous catalytic binding site. Inhibition of CYP3A4 often leads to undesired drug–drug interactions and toxic side effects. Previous studies have demonstrated that enantiomers of the same chiral drug might inhibit CYP3A4 with different potencies and thus, one enantiomer might have different pharmacokinetics, and metabolism as compared to its counterpart. Therefore, the use of enantiopure therapeutic agents has been advocated as a promising concept to evade drug toxicity. This emphasizes the need to investigate the molecular basis of interaction and stereoselectivity of CYP3A4. Towards this goal, various in silico models have been developed that correlate the differences in the inhibitory potencies of enantiomeric pairs with difference in their interaction pattern within the binding cavity of CYP3A4. The 3D interaction based differentiating features include two hydrogen bond acceptors at a certain distance from each other, as well as from molecular boundaries of different enantiomeric pairs. Additionally, our results demonstrated the role of a hydrogen bond donor and a hydrogen bond acceptor in stereoselective inhibition of CYP3A4.
Graphical Abstract
Open image in new window
For this type of study formal consent is not required
... The essential nonbonded interactions were represented by the probes: DRY (hydrophobic interactions), O (hydrogen bond acceptor), N1 (hydrogen bond donor), and TIP (molecular shape descriptor). GRID with step 0.5 was the method for MIF computation, while the discretization algorithm was AMANDA [48] with a scale factor of 0.55. The encoding method was MACC2 with weights set to DRY: −0.5, O: −2.6, N1: −4.2, and TIP: −0.75. ...
The influenza A virus nonstructural protein 1 (NS1), which is crucial for viral replication and immune evasion, has been identified as a significant drug target with substantial potential to contribute to the fight against influenza. The emergence of drug-resistant influenza A virus strains highlights the urgent need for novel therapeutics. This study proposes a combined theoretical criterion for the virtual screening of molecular libraries to identify candidate NS1 inhibitors. By applying the criterion to the ZINC Natural Product database, followed by ligand-based virtual screening and molecular docking, we proposed the most promising candidate as a potential NS1 inhibitor. Subsequently, the selected natural compound was experimentally evaluated, revealing measurable virus replication inhibition activity in cell culture. This approach offers a promising avenue for developing novel anti-influenza agents targeting the NS1 protein.
... GRIND descriptors were calculated at grid resolution of 0.5 Å with the smoothing window 0.8 Å. At each point, the interaction energy was calculated as a sum of Lennard-Jones energy, hydrogen bond, and electrostatic interactions [34]. The DRY, N 1 , O, and TIP probes were employed to generate MIFs. ...
In this work, three quantitative structure–activity relationship (QSAR) methods involving comparative molecular field analysis (CoMFA) and GRid-INdependent Descriptors (GRIND) based 3D-QSAR and hologram QSAR (HQSAR) were evaluated for predicting adsorption coefficients of the simple benzene derivatives on mutiwalled carbon nanotubes (MWCNTs). The contour maps of CoMFA suggested that the steric hindrance had a significant impact on the adsorption process of substituted benzenes. GRIND studies investigate the important mutual distances between molecular features, which confirmed the role of hydrophobic groups as well as their distances from different steric hot spots in the benzene ring of the molecules. According to HQSAR model and its fragment contribution map, the hydrogen bond donor and acceptor were also found to play an important role in governing adsorption of substituted benzenes on CNTs. The CoMFA, GRIND, and HQSAR methods employed to build predictive 2D- and 3D-QSAR models for adsorption of simple benzene derivatives on CNTs in aqueous media successfully.
... The virtual screening of compound candidates was carried by the use of the shortest centroid distance. The calculation was carried in Pentacle software version 1.06 for Linux [77][78][79]. From each of the three candidate groups, ten of the most similar molecules were selected and further submitted to molecular docking. ...
Alzheimer’s disease (AD), a devastating neurodegenerative disease, is the focus of pharmacological research. One of the targets that attract the most attention for the potential therapy of AD is the serotonin 5HT6 receptor, which is the receptor situated exclusively in CNS on glutamatergic and GABAergic neurons. The neurochemical impact of this receptor supports the hypothesis about its role in cognitive, learning, and memory systems, which are of critical importance for AD. Natural products are a promising source of novel bioactive compounds with potential therapeutic potential as a 5HT6 receptor antagonist in the treatment of AD dementia. The ZINC—natural product database was in silico screened in order to find the candidate antagonists of 5-HT6 receptor against AD. A virtual screening protocol that includes both short-and long-range interactions between interacting molecules was employed. First, the EIIP/AQVN filter was applied for in silico screening of the ZINC database followed by 3D QSAR and molecular docking. Ten best candidate compounds were selected from the ZINC Natural Product database as potential 5HT6 Receptor antagonists and were proposed for further evaluation. The best candidate was evaluated by molecular dynamics simulations and free energy calculations.
... The most favorable regions from MIFs were extracted using AMANDA node -algorithm based on the field intensity at each node of MIFs and also mutual node MACC2) algorithm ( correlation -and cross . Finally, the maximum auto 21 interaction distance . The remnant and encoded MIFs were considered as the 22 was used for encoding the MIFs olecular descriptors (GRIND) and correlated with the experimentally independent m -GRID Partial least square SAR model. ...
Purpose: Drug repurposing is an approach successfully used for discovery of new therapeutic applications for the existing drugs. The current study was aimed to use the combination of in silico methods to identify FDA-approved drugs with possible S1P1 agonistic activity useful in multiple sclerosis (MS).
Methods: For this, a 3D-QSAR model for the known 21 S1P1 agonists were generated based on 3D-QSAR approach and used to predict the possible S1P1 agonistic activity of FDA-approved drugs. Then, the selected compounds were screened by docking into S1P1 and S1P3 receptors to select the S1P1 potent and selective compounds. Further evaluation was carried out by molecular dynamics (MD) simulation studies where the S1P1 binding energies of selected compounds were calculated.
Results: The analyses resulted in identification of cobicistat, benzonatate and brigatinib as the selective and potent S1P1 agonists with the binding energies of -85.93, -69.77 and -67.44 kcal. mol⁻¹, calculated using MM-GBSA algorithm based on 50 ns MD simulation trajectories. These values are better than that of siponimod (-59.35 kcal mol⁻¹), an FDA approved S1P1 agonist indicated for MS treatment. Furthermore, similarity network analysis revealed that cobicistat and brigatinib are the most structurally favorable compounds to interact with S1P1.
Conclusion: The findings in this study revealed that cobicistat and brigatinib can be evaluated in experimental studies as potential S1P1 agonist candidates useful in the treatment of MS.
... The program takes the matrix of interaction energies (IE) and converts it into a set of variables, where each variable represents the maximum product of two IE at the defined distance. The most relevant positions for each MIF type were extracted using the AMANDA discretization algorithm, 32 and the selected hot-spots were encoded into GRIND descriptors using the maximum autoand cross-correlation (MACC2) algorithm. After the application of one cycle of fractional factorial design (FFD) for the removal of redundant variables, 388 variables were retained in the model. ...
Antimicrobial-resistance (AMR) has become the greatest concern and highly challenging issue when treating nosocomial infections. The exigency to develop new potent compounds continues to increase worldwide, whereby derivatives of natural products are becoming more attractive. In the present paper, the microbiological assessment of a series of 12 cinnamide hydrazides, four of them completely novel, against clinically relevant pathogens has discovered several derivatives with promising in vitro activities against Acinetobacter baumannii, one of the most dreaded opportunistic pathogens in hospitals. The compounds were synthesized by combining one of three different natural acids (cinnamic, 4-chloro or 4-methoxy) with four monothiocarbohydrazones (MTCHs) - an important class of synthetic organic molecules. Their structure was confirmed by elemental microanalysis, as well as ATR-FTIR, ¹H and ¹³C NMR spectra, with the addition of 2D NMR spectra for novel compounds. The hybrids of cinnamic acids and pyridine derivatives are particularly active compounds with the lowest MIC50 value of 10.4 for p-chloro cinnamic acid and acetyl pyridine derivatives. An alignment-independent 3D QSAR model identified pharmacophoric hotspots and suggested several structural modifications that might improve the potency of this class of compounds against A. baumannii. The compounds are strong iron-chelating agents forming complexes with a stability constant between 10⁷ and 10⁹. The synthesized derivatives represent a promising class of antibacterial compounds with activities comparable to the commonly used antibiotics.
... AMANDA algorithm was used to extract the most relevant and significant MIFs along with evaluation of structural characteristics of the dataset explained by GRIND descriptors (Durán, Martínez, and Pastor 2008). The default GRID space of 0.5 and the energy cutoff values, which are -4.2, ...
Leukotrienes (LTs) are pro-inflammatory lipid mediators derived from arachidonic acid (AA), and their high production has been reported in multiple allergic, autoimmune, and cardiovascular disorders. The biological synthesis of leukotrienes is instigated by transfer of AA to 5-lipoxygenase (5-LO) via the 5-lipoxygenase-activating protein (FLAP). Suppression of FLAP can inhibit LT production at the earliest level, providing relief to patients requiring anti-leukotriene therapy. Over the last 3 decades, several FLAP modulators have been synthesized and pharmacologically tested, but none of them could be able to reach the market. Therefore, it is highly desirable to unveil the structural requirement of FLAP modulators. Here, in this study, supervised machine learning techniques and molecular modeling strategies are adapted to vaticinate the important 2D and 3D anti-inflammatory properties of structurally diverse FLAP inhibitors, respectively. For this purpose, multiple machine learning classification models have been developed to reveal the most relevant 2D features. Furthermore, to probe the 3D molecular basis of interaction of diverse anti-inflammatory compounds with FLAP, molecular docking studies were executed. By using the most probable binding poses from docking studies, the GRIND model was developed, which indicated the positive contribution of four hydrophobic, two hydrogen bond acceptor, and two shape-based features at certain distances from each other towards the inhibitory potency of FLAP modulators. Collectively, this study sheds light on important two-dimensional and three-dimensional structural requirements of FLAP modulators that can potentially guide the development of more potent chemotypes for the treatment of inflammatory disorders.
... Most relevant regions were extracted using a built-in algorithm AMANDA (Dur an et al., 2008). Default cutoff values for probes were used to discretize the MIFs. ...
Insulin like growth factor receptor (IGF-1R) and Insulin receptor (IR) are widely accepted to play a prominent role in cancer drug discovery due to their well-established involvement in various stages of tumorigenesis. Previously, neutralization of IGF-1R via monoclonal antibodies was in focus, which failed because of compensatory activation of IR-A upon inhibition of IGF-1R. Recent studies have demonstrated high homology between IGF-IR and IR particularly in tyrosine kinase domain and targeting both receptors have produced efficient therapeutic approaches such as inhibition of cancer cell cycle proliferation. Herein, we have made an attempt to analyze the unique data set from different chemical classes, containing potent ATP competitors against tyrosine kinase domain. We performed the 2D, 3D quantitative structure-activity relationship (QSAR) studies on inhibitors of these receptors to predict useful pharmacophoric features. We have optimized virtual screening of structurally diverse data set of dual inhibitors of IGF-1R and IR. Based on QSAR studies, we predict potential novel clinical candidates with a demonstrated absorption, distribution, metabolism, elimination, and toxicology (ADMETox) track. We also demonstrated comprehensive analysis of co-crystal complexes along with their inhibitors and built 3D- GRid INdependent Descriptors (GRIND) model to obtain insightful features such as H-bond donors and acceptors, overall topology and Vander Waal volume (vdw_vol) which are found to be responsible for dual inhibition of receptors. These findings lead to further description that Tirofiban, Practolol, Edoxaban, Novobiocin have potential to perform dual inhibition of both targets.
... Molecular interaction field (MIF) calculations were performed by placing each probe at different GRID steps iteratively. Furthermore, total interaction energy (E xyz ) as a sum of Lennard-Jones potential energy (E lj ), electrostatic (E el ) potential interactions, and hydrogen-bond (E hb ) interactions was calculated at each grid point as shown in Equation (6) [134,135]: E xyz = ∑ E lj + ∑ E el + ∑ E hb (6) The most significant MIFs calculated were selected by the AMANDA algorithm [136] for the discretization step based upon the distance and the intensity value of each node (ligand-protein complex) probe. Default energy cutoff values (−0.75, −0.5, −2.6, and −4.2 for Tip, Dry, O, and N1 probes, respectively) were used for the discretization of MIFs. ...
Inositol 1, 4, 5-trisphosphate receptor (IP3R)-mediated Ca2+ signaling plays a pivotal role in different cellular processes, including cell proliferation and cell death. Remodeling Ca2+ signals by targeting the downstream effectors is considered an important hallmark in cancer progression. Despite recent structural analyses, no binding hypothesis for antagonists within the IP3-binding core (IBC) has been proposed yet. Therefore, to elucidate the 3D structural features of IP3R modulators, we used combined pharmacoinformatic approaches, including ligand-based pharmacophore models and grid-independent molecular descriptor (GRIND)-based models. Our pharmacophore model illuminates the existence of two hydrogen-bond acceptors (2.62 Å and 4.79 Å) and two hydrogen-bond donors (5.56 Å and 7.68 Å), respectively, from a hydrophobic group within the chemical scaffold, which may enhance the liability (IC50) of a compound for IP3R inhibition. Moreover, our GRIND model (PLS: Q2 = 0.70 and R2 = 0.72) further strengthens the identified pharmacophore features of IP3R modulators by probing the presence of complementary hydrogen-bond donor and hydrogen-bond acceptor hotspots at a distance of 7.6–8.0 Å and 6.8–7.2 Å, respectively, from a hydrophobic hotspot at the virtual receptor site (VRS). The identified 3D structural features of IP3R modulators were used to screen (virtual screening) 735,735 compounds from the ChemBridge database, 265,242 compounds from the National Cancer Institute (NCI) database, and 885 natural compounds from the ZINC database. After the application of filters, four compounds from ChemBridge, one compound from ZINC, and three compounds from NCI were shortlisted as potential hits (antagonists) against IP3R. The identified hits could further assist in the design and optimization of lead structures for the targeting and remodeling of Ca2+ signals in cancer.
... Grid step of 0.5 A was used to sample the box enclosing the molecules. The most important regions that represent favorable interactions between the probe and the ligand were separated by using the ALMOND algorithm [22]. The CLACC (consistently large auto-and cross-correlation) methodology [23] was used for the encoding. ...
HIV protease inhibitors are one of the most important agents for the treatment of HIV infection. In this work, molecular modeling studies combining 3D-QSAR, molecular docking, MESP, HOMO, and LUMO energy calculations were performed on propiophenone derivatives to explore structure activity relationships and structural requirements for the inhibitory activity. The aim of this study was to create a field point–based 3D-QSAR (3D-Quantitative structure-activity relationship) model by using chalcone structures with anti-HIV-1 protease activity from our previous study and to design new potentially more potent and safer inhibitors. The developed model showed acceptable predictive and descriptive capability as represented by standard statistical parameters R2 (0.94) and Q2 (0.59). High correlation between experimental and predicted activities of training set is noticed. All compounds fit into the defined applicability domain. The derived pharmacophoric features were further supported by MESP and Mulliken charge analysis using density functional theory. Statistically significant variables from 3D-QSAR were used to define key structural characteristics which enhance anti-HIV-1 protease activity. This information has been used to design new structures with anti-HIV-1 protease activity. Docking studies were conducted to understand the interactions in predicted compounds. All the compounds were subjected to in silico ADMET profiling in order to select the best potential drug candidates.
... All MIFs were computed with the grid resolution of 0.5 Å with the smoothing window 1 Å. For the extraction of nodes from the obtained MIF AMANDA algorithm [17] was used. The distance and relative position of nodes were described by MACC2. ...
A set of solvents were classified into the switchable-hydrophilicity solvents (SHSs) and non-switchable-hydrophilicity solvents based on forming or not forming a biphasic mixture with water. SHSs have been developed to make the reaction and product separation processes easier. Herein, three classifier algorithms and various feature selection techniques relay on 3D-molecular descriptors to characterize chemicals and forecast their classes were employed. Cfs-SVM method was employed to perform a classification study. The importance of this study helps to understand more about the presence of hydrophobic groups, their position, and their shape in the molecule.
... Besides b-carbolines, cyclic imides containing the phthalimide, maleimide or succinimide subunits, have been described as growth inhibitors of M. tuberculosis [16e20]. In these studies, a series of phthalimide derivatives evaluated against M. tuberculosis H 37 Rv, showed to be effective against the bacillus with minimum inhibitory concentration (MIC) ranging from 3.9 to 12.5 mg/mL [17,18,20], and were considered new lead compounds to be studied for the treatment of susceptible and multidrug resistant tuberculosis [20]. Taking into account, the anti-M. ...
A series of methyl β-carboline carboxylates (2a-g) and of imide-β-carboline derivatives containing the phthalimide (4a-g), maleimide (5b, g) and succinimide (6b, e, g) moiety were synthesized, and evaluated for their activity against Mycobacterium tuberculosis H37Rv. The most active β-carboline derivatives against the reference strain were assayed for their cytotoxicity and the activity against resistant M. tuberculosis clinical isolates. Farther, structure-activity relationship (SAR) studies were carried out using the three and four-dimensional approaches for starting to understand the way of β-carboline activity in M. tuberculosis. All 19 β-carboline derivatives were assayed, firstly, by determining the minimum inhibitory concentration (MIC) using resazurin microtiter assay plate (REMA) in M. tuberculosis H37Rv. Then, five derivatives (2c, 4a, 4e, 4g, 6g), which showed MIC ≤ 125 μg/mL, were assayed in nine resistant M. tuberculosis clinical isolates (five MDR, three isoniazid monoresistant and one isoniazid plus streptomycin resistant). The MIC values against the resistant clinical isolates ranged from 31.25 to >250 μg/mL. All five derivatives were non-cytotoxic to the VERO cell line, determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, at the tested concentration (selectivity index ranged from <1.74 to 14.4). Our study demonstrated that (2c) and (6g) derivatives had better anti-M. tuberculosis activity, especially against resistant clinical isolates, what makes them scaffold candidates for further investigations about their anti-tuberculosis activity. The SAR study conducted with the 19 β-carboline derivatives showed the importance of steric effects for the synthesized β-carbolines against M tuberculosis, and these models can be used for future proposition of new derivatives, increasing the chances of obtaining potentially anti-tuberculosis compounds.
... All MIFs were computed with the grid resolution of 0.5 Å and the smoothing window 1 Å. For the extraction of nodes from the MIF obtained, AMANDA [12] algorithm was used and the distance and relative position of nodes were described by MACC2. The GRIND is a function of the distance instead of the position of each grid point. ...
The control of permeation is vital not only for the topical application of lotions, creams, and ointments but also for the toxicological and risk assessment of materials from environmental and occupational hazards. To understand the effects of physicochemical properties of a variety of 211 compounds on skin permeability, we developed a three-dimensional quantitative structure-property relationship (3 D-QSPR) model. Alignment free GRid-INdependent Descriptors (GRINDs), which were derived from molecular interaction fields (MIFs) contributed to the regression models. Kennard-Stone algorithm was employed to split data set to a training set of 159 molecules and a test set of 52 molecules. Fractional factorial design (FFD), genetic algorithm (GA) and successive projection algorithm (SPA) were applied to select the most relevant 3 D molecular descriptors. The descriptors selected using various feature selection were correlated with skin permeability constants by partial least squares (PLS) and support vector machine (SVM). SPA-SVM model gave prominent statistical values with the correlation coefficient of [Formula: see text]= 0.96, Q2= 0.73 and R2pred=0.76. According to the analysis results, the hydrogen bonding donor and acceptor properties of the investigated compounds can influence the penetration into the human skin. Furthermore, it was found that permeability was enhanced by increasing the hydrophobicity and was diminished by increasing the molecular weight. In addition, the presence of hydrophobic groups in the target molecule, as well as their shape and position, can affect the skin permeability.
... The process was followed by discarding the nodes with the energies below the default cut-off values. Then the most favorable regions were extracted from the produced MIFs using AMANDA algorithm based on field intensity at a node and the mutual nodenode distances between the selected nodes [30]. Finally, the encoding of MIFs were processed by the Consistently Large Auto and Cross Correlation (CLACC) algorithm [31] which resulted in more consistent variables compared to Maximum Auto-And Cross-Correlation (MACC) method. ...
Sphingosine 1-phosphate type 1 (S1P1) receptors are expressed on lymphocytes and regulate immune cells trafficking. Sphingosine 1-phosphate and its analogues cause internalization and degradation of S1P1 receptors, preventing the auto reactivity of immune cells in the target tissues. It has been shown that S1P1 receptor agonists such as fingolimod can be suitable candidates for treatment of autoimmune diseases. The current study aimed to generate GRIND-based 3D-QSAR predictive models for agonistic activities of 2-imino-thiazolidin-4-one derivatives on S1P1 to be used in virtual screening of chemical libraries. The developed model for the S1P1 receptor agonists showed appropriate power of predictivity in internal (r2acc 0.93 and SDEC 0.18) and external (r2 0.75 and MAE (95% data), 0.28) validations. The generated model revealed the importance of variables DRY-N1 and DRY-O in the potency and selectivity of these compounds towards S1P1 receptor. To propose potential chemical entities with S1P1 agonistic activity, PubChem chemicals database was searched and the selected compounds were virtually tested for S1P1 receptor agonistic activity using the generated models, which resulted in four potential compounds with high potency and selectivity towards S1P1 receptor. Moreover, the affinities of the identified compounds towards S1P1 receptor were evaluated using molecular dynamics simulations. The results indicated that the binding energies of the compounds were in the range of -39.31 to -46.18 and -3.20 to -9.75 kcal mol-1, calculated by MM-GBSA and MM-PBSA algorithms, respectively. The findings in the current work may be useful for the identification of potent and selective S1P1 receptor agonists with potential use in diseases such as multiple sclerosis.
... The AMANDA algorithm selects all atoms represented in the set of nodes. A node prefiltration step is carried out at the first step by applying suitable energy cutoff values [57]. The discretization step was performed using default energy cutoff values (DRY, −0.75 kcal/mol; O, −0.5 kcal/mol; N1, −4.2 kcal/mol; and TIP, −2.6 kcal/mol) of the respective probes. ...
Human ether a-go-go related gene (hERG) or KV11.1 potassium channels mediate the rapid delayed rectifier current (IKr) in cardiac myocytes. Drug-induced inhibition of hERG channels has been implicated in the development of acquired long QT syndrome type (aLQTS) and fatal arrhythmias. Several marketed drugs have been withdrawn for this reason. Therefore, there is considerable interest in developing better tests for predicting drugs which can block the hERG channel. The drug-binding pocket in hERG channels, which lies below the selectivity filter, normally contains K+ ions and water molecules. In this study, we test the hypothesis that these water molecules impact drug binding to hERG. We developed 3D QSAR models based on alignment independent descriptors (GRIND) using docked ligands in open and closed conformations of hERG in the presence (solvated) and absence (non-solvated) of water molecules. The ligand–protein interaction fingerprints (PLIF) scheme was used to summarize and compare the interactions. All models delineated similar 3D hERG binding features, however, small deviations of about ~0.4 Å were observed between important hotspots of molecular interaction fields (MIFs) between solvated and non-solvated hERG models. These small changes in conformations do not affect the performance and predictive power of the model to any significant extent. The model that exhibits the best statistical values was attained with a cryo_EM structure of the hERG channel in open state without water. This model also showed the best R2 of 0.58 and 0.51 for the internal and external validation test sets respectively. Our results suggest that the inclusion of water molecules during the docking process has little effect on conformations and this conformational change does not impact the predictive ability of the 3D QSAR models.
... Applied probes (mapped regions of molecule surface) were DRY (hydrophobic interactions) O (hydrogen bond acceptor) N1 (hydrogen bond donor) and TIP (molecular shape descriptor). Discretization Method was AMANDA (Duran et al., 2008), with scale factor 0.55. Cut off was set to: ...
Influenza A virus (IAV) matrix protein 2 (M2), an ion channel, is crucial for virus infection, and therefore, an important anti-influenza drug target. Adamantanes, also known as M2 channel blockers, are one of the two classes of Food and Drug Administration-approved anti-influenza drugs, although their use was discontinued due to prevalent drug resistance. Fast emergence of resistance to current anti-influenza drugs have raised an urgent need for developing new anti-influenza drugs against resistant forms of circulating viruses. Here we propose a simple theoretical criterion for fast virtual screening of molecular libraries for candidate anti-influenza ion channel inhibitors both for wild type and adamantane-resistant influenza A viruses. After in silico screening of drug space using the EIIP/AQVN filter and further filtering of drugs by ligand based virtual screening and molecular docking we propose the best candidate drugs as potential dual inhibitors of wild type and adamantane-resistant influenza A viruses. Finally, guanethidine, the best ranked drug selected from ligand-based virtual screening, was experimentally tested. The experimental results show measurable anti-influenza activity of guanethidine in cell culture.
... 2. Discretization: AMANDA algorithm (Durán, Martínez & Pastor, 2008) was used to discretize the MIFs using default energy cutoff values of −0.75, −0.5, −4.2 and −2.6 for the TIP, DRY, N1 and O probes, respectively, to pre-filter the nodes that fails to meet the energy cutoffs. ...
Background
The γ-aminobutyric acid (GABA) transporter GAT1 is involved in GABA transport across the biological membrane in and out of the synaptic cleft. The efficiency of this Na ⁺ coupled GABA transport is regulated by an electrochemical gradient, which is directed inward under normal conditions. However, in certain pathophysiological situations, including strong depolarization or an imbalance in ion homeostasis, the GABA influx into the cytoplasm is increased by re-uptake transport mechanism. This mechanism may lead to extra removal of extracellular GABA which results in numerous neurological disorders such as epilepsy. Thus, small molecule inhibitors of GABA re-uptake may enhance GABA activity at the synaptic clefts.
Methods
In the present study, various GRID-independent molecular descriptor (GRIND) models have been developed to shed light on the 3D structural features of human GAT1 (hGAT1) inhibitors using nipecotic acid and N-diarylalkenyl piperidine analogs. Further, a binding hypothesis has been developed for the selected GAT1 antagonists by molecular docking inside the binding cavity of hGAT1 homology model.
Results
Our results indicate that two hydrogen bond acceptors, one hydrogen bond donor and one hydrophobic region at certain distances from each other play an important role in achieving high inhibitory potency against hGAT1. Our docking results elucidate the importance of the COOH group in hGAT1 antagonists by considering substitution of the COOH group with an isoxazol ring in compound 37 , which subsequently leads to a three order of magnitude decrease in biological activity of 37 (IC 50 = 38 µM) as compared to compound 1 (IC 50 = 0.040 µM).
Discussion
Our docking results are strengthened by the structure activity relationship of the data series as well as by GRIND models, thus providing a significant structural basis for understanding the binding of antagonists, which may be useful for guiding the design of hGAT1 inhibitors.
... The previously developed GRIND model uses alignment-independent molecular descriptors for partial least square (PLS) analysis. Briefly, GRIND uses four different probes including DRY (hydrophobic interactions), N1 (hydrogen bond acceptor), O (hydrogen bond donor), and TIP (steric hotspots) for the computation of molecular interaction fields (MIFs) of the data set of compounds (Pastor, Cruciani, McLay, Pickett & Clementi, 2000) followed by MIF discretization and encoding by AMANDA (Durán, Martínez & Pastor, 2008) and Consistently Large Auto And Cross Correlation (CLACC) algorithm, respectively a into alignment-independent variables for partial least square (PLS) analysis. In the current study, selected hits were treated as test set for the prediction of pIC 50 values against PH domain of Akt2 using GRIND-based PLS model. ...
Protein Kinase B‐beta (PKBβ/Akt2) is a non‐receptor kinase that has attracted a great deal of attention as a promising cancer therapy drug target. In mammalian cells, hyper‐activation of Akt2 exclusively facilitates the survival of solid tumors by interfering with cell cycle progression. This definite function of Akt2 in tumor survival/maintenance provides the basis for the development of its antagonists with the aim of desensitizing cell proliferation. In order to find novel and potent Akt2 inhibitors, structure‐based pharmacophore models have been developed and validated by the test set prediction. The final pharmacophore model was used for hits identification using public chemical databases. The hits were further prioritized using drug‐like filters which revealed 14 potential hit compounds having novel chemical scaffolds. Our results elucidate the importance of three hydrogen bond acceptors (A), one hydrogen bond donor (D), one hydrophobic group (H) and one positive ionic charge (P) towards inhibition of the Ak2. One of our selected hits showed 68% cell apoptosis at 8 μg/mL concentration. We proposed various chemical scaffolds including benzamide, carboxamide, and methyl benzimidazole targeting Akt2 and thus may act as potential leads for the further development of new anticancer agents.
This article is protected by copyright. All rights reserved.
... Our hope was not unfounded, and different similar methods have been published in the past (Tetko et al., 2005;Masek et al., 2008). For this particular purpose, we obtained excellent results using a simple random permutation of the molecular descriptors generated by methods like GRIND or GRIND2 (Pastor et al., 2000;Durán et al., 2008). The permuted vector of descriptors does not allow guessing the original structure, since the permutation destroyed any link between the value of the variables and their physicochemical interpretation. ...
In silico methods are increasingly being used for assessing the chemical safety of substances, as a part of integrated approaches involving in vitro and in vivo experiments. A paradigmatic example of these strategies is the eTOX project http://www.etoxproject.eu, funded by the European Innovative Medicines Initiative (IMI), which aimed at producing high quality predictions of in vivo toxicity of drug candidates and resulted in generating about 200 models for diverse endpoints of toxicological interest. In an industry-oriented project like eTOX, apart from the predictive quality, the models need to meet other quality parameters related to the procedures for their generation and their intended use. For example, when the models are used for predicting the properties of drug candidates, the prediction system must guarantee the complete confidentiality of the compound structures. The interface of the system must be designed to provide non-expert users all the information required to choose the models and appropriately interpret the results. Moreover, procedures like installation, maintenance, documentation, validation and versioning, which are common in software development, must be also implemented for the models and for the prediction platform in which they are implemented. In this article we describe our experience in the eTOX project and the lessons learned after 7 years of close collaboration between industrial and academic partners. We believe that some of the solutions found and the tools developed could be useful for supporting similar initiatives in the future.
... Partial atomic charges have been calculated by Gasteiger-Huckel method. Structure-independent descriptors, known as GRINDs (Grid Independent Descriptors), were afterward calculated for chemical compounds using AMANDA algorithm [33,34]. AMANDA algorithm works through following steps: (i) calculating molecular interaction fields (MIFs) followed by identifying the most favorable interaction energies, termed nodes; (ii) node filtration to reach a set of regions with the most favorable interaction energies. ...
Nowadays, antibiotic resistance has turned into one of the most important worldwide health problems. Biological end point of critical enzymes induced by potent inhibitors is recently being considered as a highly effective and popular strategy to defeat antibiotic-resistant pathogens. For instance, the simple but critical β-carbonic anhydrase has recently been in the center of attention for anti-pathogen drug discoveries. However, no β-carbonic anhydrase selective inhibitor has yet been developed. Available β-carbonic anhydrase inhibitors are also highly potent with regard to human carbonic anhydrases, leading to severe inevitable side effects in case of usage. Therefore, developing novel inhibitors with high selectivity against pathogenic β-carbonic anhydrases is of great essence. Herein, for the first time, we have conducted a proteochemometric study to explore the structural and the chemical aspects of the interactions governed by bacterial β-carbonic anhydrases and their inhibitors. We have found valuable information which can lead to designing novel inhibitors with better selectivity for bacterial β-carbonic anhydrases.
... MIFs were calculated using four chemical probes: DRY probe, which maps the hydrophobic regions of a molecule; O is sp 2 carbonyl probe that discovers the regions of a molecule with H-bond donor ability; N1 is neutral flat -NH that maps the H-bond acceptors of molecules, and TIP probe encodes the shape of a molecule. Important positions around molecules (hot spots) are extracted from MIFs using AMANDA discretization algorithm [34]. Encoding of the filtered MIFs into GRIND variables was performed by the maximal auto-and cross-correlation (MACC2) algorithm. ...
The ratios of E/Z isomers of sixteen synthesized 1,3-dihydro-3-(substituted
phenylimino)-2H-indol-2-one were studied using experimental and theoretical methodology.
Linear solvation energy relationships (LSER) rationalized solvent influence of the solvent–solute
interactions on the UV-Vis absorption maxima shifts (νmax) of both geometrical isomers using the
Kamlet-Taft equation. Linear free energy relationships (LFER) in the form of single substituent
parameter equation (SSP) was used to analyze substituent effect on pKa
, NMR chemical shifts and
νmax values. Electron charge density was obtained by the use of Quantum Theory of Atoms in
Molecules, i.e. Bader's analysis. The substituent and solvent effect on intramolecular charge
transfer (ICT) were interpreted with the aid of time-dependent density functional (TD-DFT)
method. Additionally, the results of TD-DFT calculations quantified the efficiency of ICT from
the calculated charge-transfer distance (DCT) and amount of transferred charge (QCT). The
antimicrobial activity was evaluated using broth microdilution method. 3D QSAR modeling was
used to demonstrate the influence of substituents effect as well as molecule geometry on
antimicrobial activity.
... Important positions around molecules (hot spots) are extracted from MIFs using AMANDA discretization algorithm [30]. Encoding of the filtered MIFs into GRIND variables was performed by the maximal auto-and cross-correlation (MACC2) algorithm. ...
Antimicrobial resistance (AMR) is a major health problem worldwide, because of ability of bacteria, fungi and viruses to evade known therapeutic agents used in treatment of infections. Aryldiketo acids (ADK) have shown antimicrobial activity against several resistant strains including Gram-positive Staphylococcus aureus bacteria. Our previous studies revealed that ADK analogues having bulky alkyl group in ortho position on a phenyl ring have up to ten times better activity than norfloxacin against the same strains. Rational modifications of analogues by introduction of hydrophobic substituents on the aromatic ring has led to more than tenfold increase in antibacterial activity against multidrug resistant Gram positive strains. To elucidate a potential mechanism of action for this potentially novel class of antimicrobials, several bacterial enzymes were identified as putative targets according to literature data and pharmacophoric similarity searches for potent ADK analogues. Among the seven bacterial targets chosen, the strongest favorable binding interactions were observed between most active analogue and S. aureus dehydrosqualene synthase and DNA gyrase. Furthermore, the docking results in combination with literature data suggest that these novel molecules could also target several other bacterial enzymes, including prenyl-transferases and methionine aminopeptidase. These results and our statistically significant 3D QSAR model could be used to guide the further design of more potent derivatives as well as in virtual screening for novel antibacterial agents.
... The structural characteristics of the dataset explained by the GRIND descriptors were evaluated with the AMANDA algorithm [71], which extracts the regions with the most relevant MIFs. The default grid space (0.5) and the energy cut-off values for probes implemented in software Pentacle v 1.07 [70] were used to discretize the MIFs. ...
Protein kinase B-β (PKBβ/Akt2) is a serine/threonine-specific protein kinase that has emerged as one of the most important regulators of cell growth, differentiation, and division. Upregulation of Akt2 in various human carcinomas, including ovarian, breast, and pancreatic, is a well-known tumorigenesis phenomenon. Early on, the concept of the simultaneous administration of anticancer drugs with inhibitors of Akt2 was advocated to overcome cell proliferation in the chemotherapeutic treatment of cancer. However, clinical studies have not lived up to the high expectations, and several phase II and phase III clinical studies have been terminated prematurely because of severe side effects related to the non-selective isomeric inhibition of Akt2. The notion that the sequence identity of pleckstrin homology (PH) domains within Akt-isoforms is less than 30% might indicate the possibility of the development of selective antagonists against the Akt2 PH domain. Therefore, in this study, various in silico tools were utilized to explore the hypothesis that quinoline-type inhibitors bind in the Akt2 PH domain. A Grid-Independent Molecular Descriptor (GRIND) analysis indicated that two hydrogen bond acceptors, two hydrogen bond donors and one hydrophobic feature at a certain distance from each other were important for the selective inhibition of Akt2. Our docking results delineated the importance of Lys30 as an anchor point for mapping the distances of important amino acid residues in the binding pocket, including Lys14, Glu17, Arg25, Asn53, Asn54 and Arg86. The binding regions identified complement the GRIND-based pharmacophoric features.
... The regions with most relevant MIFs were extracted using AMANDA algorithm [16] implemented in the Pentacle. Default cutoff values for probes energies were used to discretize the MIFs. ...
ATP-dependent xenobiotic efflux transporter P-glycoprotein (P-gp) limits the cellular accumulation of many therapeutically important drug molecules. The most prominent of these are CNS active compounds and the potential chemotherapeutic agents. Co-administration of chemotherapeutic agents with modulators of P-glycoprotein has been advocated as a promising concept to circumvent drug resistance in tumor cells. Several pharmacoinformatics strategies to investigate ligand-P-glycoprotein interactions profiles revealed that these are promiscuous, multi-site and conformational dependent processes that take place in asymmetric 3D space within the binding cavity of P-gp. Therefore, avoiding stereoselectivity associated with ligand-protein interaction may compromise efficiency of the QSAR and other modeling strategies. Within this article, several SAR and QSAR strategies in combination with molecular docking studies on stereoisomeric inhibitors of P-gp will be highlighted to further explore the stereoselectivity of ligand-P-glycoprotein interaction.
... The regions with most relevant MIFs were extracted using AMANDA algorithm [16] implemented in the Pentacle. Default cutoff values for probes energies were used to discretize the MIFs. ...
Background:
ATP-binding cassette (ABC) transporters, P-glycoprotein (P-gp, ABCB1) and breast cancer resistance protein (BCRP/ABCG2) are major determinants of pharmacokinetic, safety and efficacy profiles of drugs thereby effluxing a broad range of endogenous substances across the plasma membrane. Overexpression of these transporters in various tumors is also implicated in the development of multidrug resistance (MDR) and thus, hampers the success of cancer chemotherapy. Modulators of these efflux transporters in combination with chemotherapeutics could be a promising concept to increase the effective intracellular concentration of anticancer drugs. However, broad and overlapped specificity for substrates and modulators of ABCB1 and ABCG2, merely induce toxicity and unwanted drug-drug interactions and thus, lead to late-stage failure of drugs.
Objective:
In present investigation, we aim to identify specific 3D structural requirements for selective inhibition of ABCB1 and ABCG2 transport function.
Method:
GRID Independent Molecular Descriptor (GRIND) models of selective inhibitors of both transporters have been developed, using their most probable binding conformations obtained from molecular docking protocol.
Results:
Our results demonstrated a dominant role of molecular shape and different H-bonding patterns in drug-ABCB1/ABCG2 selective interactions. Moreover, distinct distances of different pharmacophoric features from steric hot spots of the molecules provided a strong basis of selectivity for both transporters. Additionally, our results suggested the presence of two H-bond donors at a distance of 8.4-8.8 Å in selective modulators of ABCG2.
Conclusion:
Our findings concluded that molecular shape along with three dimensional pattern of Hbonding in MDR modulators play a critical role in determining the selectivity between the two targets.
... The consensus docking pose found for the initial ligand series was governed by the double hydrogen bond of Asn250 6.55 via the pyrimidine N atom and the exocyclic amine, combined with hydrophobic packing between with Phe168 EL2 and Leu246 6.51 . This docking pose was used as a molecular alignment to generate the 3D-QSAR model, produced with the second generation of GRid INdependent Descriptors (GRIND-2) [78]. The results of this modeling were combined with comparative sequence and structural analysis of the binding site residues to elucidate the reasons of the high A 3 AR selectivity. ...
The family of adenosine receptors (ARs) is focus of several medicinal chemistry programs aimed to find new potent and selective drugs. Each receptor subtype has been proposed as a relevant drug target in the treatment of, e.g., cardiovascular or inflammatory diseases, asthma or Parkinson's disease. Until recently, most of these efforts have been dominated by ligand-based or empirical approaches. However, the latest advances in G protein-coupled receptor (GPCR) crystallography allowed for a thorough structural characterization of the A2AAR subtype, which has been crystalized with a number of agonists and antagonists. Consequently, the ligand discovery of AR ligands has been enriched with a number of structure-based approaches. These include the generation of higher-confident homology models for the remaining AR subtypes, virtual screening identification of novel chemotypes, structure-based lead-optimization programs, rationalization of selectivity profiles, or the structural characterization of novel binding sites that enable the design of novel allosteric modulators. Computational methodologies have importantly contributed to the success of these structure-based approaches, and the recent advances in the field are also analyzed in this review. We conclude that the design of adenosine receptor ligands has improved dramatically with the consideration of structure- based approaches, which is paving the way to a better understanding of the biology and pharmacological modulation of this relevant family of receptors.
... using Tripos force field with a distance-dependent dielectric and the Powell conjugate gradient algorithm convergence criterion of 0.01 kcal/mol A. The Gasteiger-Huckel method was used to calculate partial atomic charges of all compounds. The final structures were subjected to PENTACLE 1.05 software (62), and gridindependent descriptors (GRINDs) (63) were calculated for them using both ALMOND and AMANDA algorithms (64). ...
The critical role of carbonic anhydrases in different physiological processes has put this protein family at the center of attention, challenging major diseases like glaucoma, neurological disorders such as epilepsy and Alzheimer's disease, obesity and cancers. Many QSAR/QSPR (quantitative structure activity/property relationship) researches have been done to design potent carbonic anhydrase inhibitors (CAIs); however using inhibitors with no selectivity for different isoforms can lead to major side effects. Given that QSAR/QSPR methods are not capable of covering multiple targets in a unified model, we have applied the proteochemometrics approach to model the interaction space that governs selective inhibition of different CA isoforms by some mono-/dihydroxybenzoic acid esters. Internal and external validation methods showed that all models were reliable in terms of both validity and predictivity, whereas Y-scrambling assessed the robustness of the models. To prove the applicability of our models we showed how structural changes of a ligand can affect the selectivity. Our models provided interesting information that can be useful for designing inhibitors with selective behavior towards isoforms of carbonic anhydrases, aiding in their selective inhibition. This article is protected by copyright. All rights reserved.
... Molecular weight (MW), MACCS Fingerprints (Accelrys), MORGAN Fingerprints with radius 2 (Morgan 1965;Rogers and Hahn 2010), and Murcko scaffolds (Bemis and Murcko 1996) were generated with RDKit. GRid INdependent Descriptors of second generation (GRIND-2) (Pastor et al. 2000;Pastor 2006) were generated using Pentacle version 1.06 (Durán et al. 2008;Durán and Pastor 2010) with default parameters and used without scaling, as recommended by the authors. ...
Most computational methods used for the prediction of toxicity endpoints are based on the assumption that similar compounds have similar biological properties. This principle can be exploited using computational methods like read across or quantitative structure-activity relationships. However, there is no general agreement about which method is the most appropriate for quantifying compound similarity neither for exploiting the similarity principle in order to obtain reliable estimations of the compound properties. Moreover, optimal similarity metrics and modeling methods might depend on the characteristics of the endpoints and training series used in each case. This study describes a comparative analysis of the predictive performance of diverse similarity metrics and modeling methods in toxicological applications. A collection of two quantitative (n = 660, n = 1114) and three qualitative (n = 447, n = 905, n = 1220) datasets representing very different endpoints of interest in drug safety evaluation and rigorous methods were used to estimate the external predictive ability in each case. The results confirm that no single approach produces the best results in all instances, and the best predictions were obtained using different tools in different situations. The trends observed in this study were exploited to propose a unifying strategy allowing the use of the most suitable method for every compound. A comparison of the quality of the predictions obtained by the unifying strategy with those obtained by standard prediction methods confirmed the usefulness of the proposed approach.
Recepteur d’Origine Nantais known as RON is a member of the receptor tyrosine kinase (RTK) superfamily which has recently gained increasing attention as cancer target for therapeutic intervention. The aim of this work was to perform an alignment-independent three-dimensional quantitative structure–activity relationship (3D QSAR) study for a series of RON inhibitors. A 3D QSAR model based on GRid-INdependent Descriptors (GRIND) methodology was generated using a set of 19 compounds with RON inhibitory activities. The generated 3D QSAR model revealed the main structural features important in the potency of RON inhibitors. The results obtained from the presented study can be used in lead optimization projects for designing of novel compounds where inhibition of RON is needed.
Background
Inflammation is common pathogenesis of many diseases progression, such as malignancy, cardiovascular and rheumatic diseases. The inhibition of the synthesis of inflammatory mediators by modulation of cyclooxygenase (COX) and lipoxygenase (LOX) pathways provides a challenging strategy for the development of more effective drugs.
Objective
The aim of this study was to design dual COX-2 and 5-LOX inhibitors with iron-chelating properties using a combination of ligand-based (three-dimensional quantitative structure-activity relationship (3D-QSAR)) and structure-based (molecular docking) methods.
Methods
The 3D-QSAR analysis was applied on a literature dataset consisting of 28 dual COX-2 and 5-LOX inhibitors in Pentacle software. The quality of developed COX-2 and 5-LOX 3D-QSAR models were evaluated by internal and external validation methods. The molecular docking analysis was performed in GOLD software, while selected ADMET properties were predicted in ADMET predictor software.
Results
According to the molecular docking studies, the class of sulfohydroxamic acid analogues, previously designed by 3D-QSAR, was clustered as potential dual COX-2 and 5-LOX inhibitors with iron-chelating properties. Based on the 3D-QSAR and molecular docking, 1j, 1g, and 1l were selected as the most promising dual COX-2 and 5-LOX inhibitors. According to the in silico ADMET predictions, all compounds had an ADMET_Risk score less than 7 and a CYP_Risk score lower than 2.5. Designed compounds were not estimated as hERG inhibitors, and 1j had improved intrinsic solubility (8.704) in comparison to the dataset compounds (0.411-7.946).
Conclusion
By combining 3D-QSAR and molecular docking, three compounds (1j, 1g, and 1l) are selected as the most promising designed dual COX-2 and 5-LOX inhibitors, for which good activity, as well as favourable ADMET properties and toxicity, are expected.
Stability constants prediction plays a critical role in the identification and optimization of ligand design for selective complexation of metal ions in solution. Thus, it is important to assess the potential of metal-binding ligand organic in the complex formation process. However, quantitative structure-activity/property relationships (QSAR/QSPR) provide a time-and cost-efficient approach to predict the stability constants of compounds. To this end, we applied a free alignment three-dimensional QSPR technique by generating GRid-INdependent Descriptors (GRINDs) to rationalize the underlying factors effecting on stability constants of transition metals; 105 (Y3+), 186 (La3+), and 66 (UO22+) with diverse organic ligands in aqueous solutions at 298 K and an ionic strength of 0.1 M. Kennard- Stone algorithm was employed to split data set to a training set of 75% molecules and a test set of 25% molecules. Fractional factorial design (FFD) and genetic algorithm (GA) applied to derive the most relevant and optimal 3 D molecular descriptors. The selected descriptors using various feature selection were correlated with stability constants by partial least squares (PLS). GA-PLS models were statistically validated (
R
2
= 0.96, q2 = 0.82 and R2pred=0.81 for Y3+;
R
2
= 0.90, q2 = 0.73 and R2pred=0.82 for La3+ and
R
2
= 0.95, q2 = 0.81 and R2pred=0.88 for UO22+), and from the information derived from the graphical results confirmed that hydrogen bonding properties, shape, size, and steric effects are the main parameters influencing stability constant of metal complexation. The provided information in this research can predict the stability constant of the new organic ligand with the transition metals without experimental processes.
In silico toxicology is an emerging field. It gains increasing importance as research is aiming to decrease the use of animal experiments as suggested in the 3R principles by Russell and Burch. In silico toxicology is a means to identify hazards of compounds before synthesis, and thus in very early stages of drug development. For chemical industries, as well as regulatory agencies it can aid in gap‐filling and guide risk minimization strategies. Techniques such as structural alerts, read‐across, quantitative structure–activity relationship, machine learning, and deep learning allow to use in silico toxicology in many cases, some even when data is scarce. Especially the concept of adverse outcome pathways puts all techniques into a broader context and can elucidate predictions by mechanistic insights.
This article is categorized under:
• Structure and Mechanism > Computational Biochemistry and Biophysics
• Data Science > Chemoinformatics
Abstract
How computers can help toxicologists' decision‐making.
A wide range of neuropsychological disorders is caused by serotonin 5-HT2A receptor (5-HT2AR) malfunction. Therefore, this receptor had been frequently used as target in CNS drug research. To design novel potent 5-HT2AR antagonists, we have combined ligand-based and target-based approaches. This study was performed on wide range of structurally diverse antagonists that were divided into three different clusters: clozapine, ziprasidone, and ChEMBL240876 derivatives. By performing the 50 ns long molecular dynamic simulations with each cluster representative in complex with 5-HT2A receptor, we have obtained virtually bioactive conformations of the ligands and three different antagonist-bound, inactive, conformations of the 5-HT2AR. These three 5-HT2AR conformations were further used for docking studies and generation of the bioactive conformations of the data set ligands in each cluster. Subsequently, selected conformers were used for 3D-Quantitative Structure Activity Relationship (3D-QSAR) modelling and pharmacophore analysis. The reliability and predictive power of the created model was assessed using an external test set compounds and showed reasonable external predictability. Statistically significant variables were used to define the most important structural features required for 5-HT2A antagonistic activity. Conclusions obtained from performed ligand-based (3D-QSAR) and target-based (molecular docking and molecular dynamics) methods were compiled and used as guidelines for rational drug design of novel 5-HT2AR antagonists.
Communicated by Ramaswamy H. Sarma
Dihydrofolate reductase (DHFR) is one of the effective and attractive chemotherapeutic targets because of its importance in catalyzing the reduction of dihydrofolate to tetrahydrofolate, which has important function in the action of folate-dependent enzymes. In this study, a new hybrid docking alignment-independent three-dimensional quantitative structure–activity relationship (3D-QSAR) model was prepared and applied in order to predict biological activities in a series of pyrimidine-based scaffold inhibitors against the human DHFR target. Partial least squares (PLS) regression was implemented to correlate genetic algorithm selected-grind descriptors with biological activity. Several internal/external approaches were used to validate the model. Considering the good agreement of all calculated parameters with the acceptance criteria and the meaningful and interpretable results of the QSAR model (i.e., PLS coefficient), we have retrieved new hits using the Pharmit server. The most active compound in the data set was docked and imported to Pharmit server in order to screen the PubChem library through building a structure-based pharmacophore model. We applied some shape filters for protein and ligand as well as Lipinski’s rule of five, in order to filter the results generated by a virtual screening which was performed by pharmacophore search. Based on the virtual screening and molecular docking results, eventually, 44 compounds were retrieved. ADMET filtration was applied to validate the hit compounds. Based on the results obtained from the in silico prediction of ADMET properties, two hits were finally retrieved from the candidate compounds.
This article describes how the Human African trypanosomiasis (HAT) is a neglected disease caused by Trypanosoma brucei. Only four drugs are available for treating HAT. Despite years of effort, little progress has been made in identifying orally available chemotypes active against the parasite. In this study, chemometric tools, such as, Principal Component Analysis (PCA) and Partial Least Squares Regression (PLS), were applied to a set of active trypanocidal N-Myristoyltransferase inhibitors. These tools were generated using the Pentacle software. The algorithm (AMANDA), the descriptors are calculated through the molecular interaction fields with the blocks: N1, O, TIP and N1, which allow extracting from them the most interesting regions. The PCA selected 507 descriptors and the scores plot clearly separated more active compounds from others. The first two PCs account for over 70% of data variance. The best PLS model, exhibits q2 = 0.762 and r2 = 0.867 and rext2 = 0.69. The results highlight the importance of acceptor hydrogen bonds regions. The importance of the metabolic cleavage of the methyl group attached to the nitrogen of pyrrole ring by N-dealkylation was observed.
Motivation: In early stage of therapeutics, several structure and ligand-based in-silico approaches have aided the modern drug discovery and design. However, such techniques are limited by availability of resolved 3D structures of targets and ligands. At the same time the growing concern of drug resistivity not only demands for new drugs but also the judicious use of presently available drugs. In such a scenario, the utilization of the already available drugs of a target molecule over the different homologous target of wider range of organisms is the better perspective for treatment. This requires confirmation of structural similarity of the targets (enzyme and protein targets) in those organisms.
Results: In the present study, based on the structural similarity of the target enzymes shared by different pathogenic micro-organisms, we have reviewed to gain an in-silico perspective of their efficacy in targeting a wider subset of organisms with few available drugs marketed for those. The results suggest efficient binding affinity of such drugs for the enzymes of organisms belonging to the cluster formed on the basis of structurally similarity.
Implementation: Such an approach can be adopted to utilize the presently available drugs for a wider range of pathogenic micro-organisms.
Supplementary Information: Available
Background: ATP-binding cassette (ABC) transporters, P-glycoprotein (P-gp, ABCB1) and breast cancer resistance protein (BCRP/ABCG2) are major determinants of pharmacokinetic, safety and efficacy profiles of drugs thereby effluxing a broad range of endogenous substances across the plasma membrane. Overexpression of these transporters in various tumors is also implicated in the development of multidrug resistance (MDR) and thus, hampers the success of cancer chemotherapy. Modulators of these efflux transporters in combination with chemotherapeutics could be a promising concept to increase the effective intracellular concentration of anticancer drugs. However, broad and overlapped specificity for substrates and modulators of ABCB1 and ABCG2, merely induce toxicity and unwanted drug-drug interactions and thus, lead to late-stage failure of drugs. Objective: In present investigation, we aim to identify specific 3D structural requirements for selective inhibition of ABCB1 and ABCG2 transport function. Method: GRID Independent Molecular Descriptor (GRIND) models of selective inhibitors of both transporters have been developed, using their most probable binding conformations obtained from molecular docking protocol. Results: Our results demonstrated a dominant role of molecular shape and different H-bonding patterns in drug-ABCB1/ABCG2 selective interactions. Moreover, distinct distances of different pharmacophoric features from steric hot spots of the molecules provided a strong basis of selectivity for both transporters. Additionally, our results suggested the presence of two H-bond donors at a distance of 8.4-8.8 A in selective modulators of ABCG2. Conclusion: Our findings concluded that molecular shape along with three dimensional pattern of H bonding in MDR modulators play a critical role in determining the selectivity between the two targets.
The knowledge of the bioactive conformation for an active hit is relevant because of the easier interpretation and the general quality of the recognition models of protein and ligand. With the aim of investigating potential bioactive conformations without previous structural knowledge of the molecular target, we present herewith a ‘protocol’ that could be used which includes generation of low-energy conformations, calculations of tridimensional descriptors and investigation of structural similarity via principal component analysis. The protocol was used in the search for potential bioactive conformations. An initial selection of targets was made from a set of protein–ligand complexes with structure deposited in the Protein Data Bank, which was systematically filtered by lead-like rules, resulting in 45 ligands of 8 important therapeutic targets. After extensive optimization of the protocol and parameters of both OMEGA and Pentacle softwares, the best results were obtained for series of compounds such as the beta-trypsin and urokinase inhibitors, which are more structurally related among each other, inside the respective therapeutic class. Future improvements of the protocol, including a suitable choice and combination of robust 3D descriptors, could yield more reliable and less restrictive results, with general and diverse applications in drug design, in particular for improving the 3D-QSAR methodologies as well as virtual screening experiments for a more reliable selection of new lead compounds for different molecular targets.
Grid-independent descriptors are extensively used in 3D quantitative structure-activity relationship (3D-QSAR) studies. These kinds of descriptors represent the spatial arrangement of the atoms in a 3D fashion. In the current study, we have developed a 3D-QSAR model for a set of hydroxamic acid-based derivatives as tumor necrosis factor-α converting enzyme (TACE) inhibitors. The generated model revealed the importance of some main moieties in the potency of these compounds. Quinolinyl and hydroxamate moieties have the ability to act as hydrogen bond acceptors and hot spot points. On the other hand, phenyl ring can participate in hydrophobic contacts with the receptor. 3D-QSAR analyses resulted in the partial least squares model of 3 latent variables with internal and external validation yielding q2 values of 0.66 and 0.73, respectively. The result of the current study can be used in designing of potent TACE inhibitors where inhibition of TACE enzyme is required, such as inflammatory diseases, atherosclerosis, osteoporosis, autoimmune disorders, allograft rejection, and cancer.
Leishmaniasis is a major group of neglected tropical diseases caused by the protozoan parasite Leishmania. About 12 million people are affected in 98 countries and 350 million people worldwide are at risk of infection. Current leishmaniasis treatments rely on a relatively small arsenal of drugs, including amphotericin B, pentamidine and others, which in general have some type of inconvenience. Recently, we have synthesized antileishmanial bis-pyridinium derivatives and symmetrical bis-pyridinium cyclophanes. These compounds are considered structural analogues of pentamidine, where the amidino moiety, protonated at physiological pH, is replaced by a positively charged nitrogen atom as a pyridinium ring.
In this work, a statistically significant GRIND2-based 3D-QSAR model was built and biological activity predictions were in silico carried out allowing rationalization of the different activities recently obtained against Leishmania donovani (in L. donovani promastigotes) for a dataset of 19 bispyridinium compounds. We will emphasize the most important structural requirements to improve the biological activity and probable interactions with the biological receptor as a guide for lead and prototype optimization. In addition, since no information about the actual biological target for this series of active compounds is provided, we have used PASS (Prediction of Activity Spectra for Biologically Active Substances) to propose our compounds as potential nicotinic α6β3β4α5 receptor antagonists. This proposal is reinforced by the high structural similarity observed between our compounds and several anthelmintic drugs in current clinical use, which have the same drug action mechanism here predicted. Such new findings would be confirmed with further and additional experimental assays.
For the successful identification and docking of new ligands to a protein target by virtual screening, the essential features of the protein and ligand surfaces must be captured and distilled in an efficient representation. Since the running time for docking increases exponentially with the number of points representing the protein and each ligand candidate, it is important to place these points where the best interactions can be made between the protein and the ligand. This definition of favorable points of interaction can also guide protein structure-based ligand design, which typically focuses on which chemical groups provide the most energetically favorable contacts. In this paper, we present an alternative method of protein template and ligand interaction point design that identifies the most favorable points for making hydrophobic and hydrogen-bond interactions by using a knowledge base. The knowledge-based protein and ligand representations have been incorporated in version 2.0 of SLIDE and resulted in dockings closer to the crystal structure orientations when screening a set of 57 known thrombin and glutathione S-transferase (GST) ligands against the apo structures of these proteins. There was also improved scoring enrichment of the dockings, meaning better differentiation between the chemically diverse known ligands and a approximately 15,000-molecule dataset of randomly-chosen small organic molecules. This approach for identifying the most important points of interaction between proteins and their ligands can equally well be used in other docking and design techniques. While much recent effort has focused on improving scoring functions for protein-ligand docking, our results indicate that improving the representation of the chemistry of proteins and their ligands is another avenue that can lead to significant improvements in the identification, docking, and scoring of ligands.
DrugBank is a richly annotated resource that combines detailed drug data with comprehensive drug target and drug action information.
Since its first release in 2006, DrugBank has been widely used to facilitate in silico drug target discovery, drug design, drug docking or screening, drug metabolism prediction, drug interaction prediction and
general pharmaceutical education. The latest version of DrugBank (release 2.0) has been expanded significantly over the previous
release. With ∼4900 drug entries, it now contains 60% more FDA-approved small molecule and biotech drugs including 10% more
‘experimental’ drugs. Significantly, more protein target data has also been added to the database, with the latest version
of DrugBank containing three times as many non-redundant protein or drug target sequences as before (1565 versus 524). Each
DrugCard entry now contains more than 100 data fields with half of the information being devoted to drug/chemical data and
the other half devoted to pharmacological, pharmacogenomic and molecular biological data. A number of new data fields, including
food–drug interactions, drug–drug interactions and experimental ADME data have been added in response to numerous user requests.
DrugBank has also significantly improved the power and simplicity of its structure query and text query searches. DrugBank
is available at http://www.drugbank.ca
For the successful identification and docking of new ligands to a protein target by virtual screening, the essential features of the protein and ligand surfaces must be captured and distilled in an efficient representation. Since the running time for docking increases exponentially with the number of points representing the protein and each ligand candidate, it is important to place these points where the best interactions can be made between the protein and the ligand. This definition of favorable points of interaction can also guide protein structure-based ligand design, which typically focuses on which chemical groups provide the most energetically favorable contacts. In this paper, we present an alternative method of protein template and ligand interaction point design that identifies the most favorable points for making hydrophobic and hydrogen–bond interactions by using a knowledge base. The knowledge-based protein and ligand representations have been incorporated in version 2.0 of SLIDE and resulted in dockings closer to the crystal structure orientations when screening a set of 57 known thrombin and glutathione S–transferase (GST) ligands against the apo structures of these proteins. There was also improved scoring enrichment of the dockings, meaning better differentiation between the chemically diverse known ligands and a ∼15,000-molecule dataset of randomly-chosen small organic molecules. This approach for identifying the most important points of interaction between proteins and their ligands can equally well be used in other docking and design techniques. While much recent effort has focused on improving scoring functions for protein-ligand docking, our results indicate that improving the representation of the chemistry of proteins and their ligands is another avenue that can lead to significant improvements in the identification, docking, and scoring of ligands.
The interaction of a probe group with a protein of known structure is computed at sample positions throughout and around the macromolecule, giving an array of energy values. The probes include water, the methyl group, amine nitrogen, carboxy oxygen, and hydroxyl. Contour surfaces at appropriate energy levels are calculated for each probe and displayed by computer graphics together with the protein structure. Contours at negative energy levels delineate contours also enable other regions of attraction between probe and protein and are found at known ligand binding clefts in particular. The contours also enable other regions of attraction to be identified and facilitate the interpretation of protein-ligand energetics. They may, therefore, be of value for drug design.
Traditional methods for performing 3D-QSAR rely upon an alignment step that is often time-consuming and can introduce user bias, the resultant model being dependent upon and sensitive to the alignment used. There are several methods which overcome this problem, but in general the necessary transformations prevent a simple interpretation of the resultant models in the original descriptor space (i.e. 3D molecular coordinates). Here we present a novel class of molecular descriptors which we have termed GRid-INdependent Descriptors (GRIND). They are derived in such a way as to be highly relevant for describing biological properties of compounds while being alignment-independent, chemically interpretable, and easy to compute. GRIND are obtained starting from a set of molecular interaction fields, computed by the program GRID or by other programs. The procedure for computing the descriptors involves a first step, in which the fields are simplified, and a second step, in which the results are encoded into alignment-independent variables using a particular type of autocorrelation transform. The molecular descriptors so obtained can be used to obtain graphical diagrams called "correlograms" and can be used in different chemometric analyses, such as principal component analysis or partial least-squares. An important feature of GRIND is that, with the use of appropriate software, the original descriptors (molecular interaction fields) can be regenerated from the autocorrelation transform and, thus, the results of the analysis represented graphically, together with the original molecular structures, in 3D plots. In this respect, the article introduces the program ALMOND, a software package developed in our group for the computation, analysis, and interpretation of GRIND. The use of the methodology is illustrated using some examples from the field of 3D-QSAR. Highly predictive and interpretable models are obtained showing the promising potential of the novel descriptors in drug design.
Cocaine is one of the most widely abused drugs in the industrial world. Substantial evidence has accumulated that the dopamine transporter (DAT) is a key target for cocaine regarding its reinforcing effects. This work describes the application of chemometric methods to a data set of 54 N(1)-benzhydryl-oxy-alkyl-N(4)-phenyl-alk(en)yl-piperazines (GBR compounds) and chemically related mepyramines as putative candidates in cocaine abuse therapy. The aim of the study is to gain insight into the structural requirements that determine the affinity of the data set molecules to the DAT and the serotonin transporter (SERT) as well as their inhibitory potency on dopamine uptake. The compounds in the dataset are described using the recently developed GRID independent descriptors (GRIND), which allow one to obtain fast three-dimensional quantitative structure-activity relationship models without the need of aligning and superimposing the structures; the results are interpreted in a convenient pharmacophoric-like fashion. In the first part of the work, the selectivity of the database molecules for DAT binding vs dopamine reuptake inhibition is investigated. In the second part, the selectivity of the compounds for DAT binding vs SERT binding is studied. In both cases, significant models are obtained, which define the structural features responsible for the respective selectivity profiles. Moreover, the information has potential interest for the design of new derivatives with improved selectivity.
This study describes the use of alignment-independent descriptors for obtaining qualitative and quantitative predictions of the competitive inhibition of CYP2C9 on a serie of highly structurally diverse compounds. This was accomplished by calculating alignment independent descriptors in ALMOND. These GRid INdependent Descriptors (GRIND) represent the most important GRID-interactions as a function of the distance instead of the actual position of each grid-point. The experimental data was determined under uniform conditions. The inhibitor data set consists of 35 structurally diverse competitive stereospecific inhibitors of the cytochrome P450 2C9 and the non -inhibitor data set of 46 compounds. In a PLS discriminant analysis 21 inhibitors and 21 non-inhibitors (1 and 0 as activities) were analyzed using the ALMOND program obtaining a model with an r
2 of 0.74 and a cross-validation value (q
2) of 0.64. The model was externally validated with 39 compounds (14 inhibitors/25 non-inhibitors). 74% of the compounds were correctly predicted and an additional 13% was assigned to a borderline cluster. Thereafter, a model for quantitative predictions was generated by a PLS analysis of the GRIND descriptors using the experimental Ki-value for 21 of the competitive inhibitors (r
2=0.77, q
2=0.60). The model was externally validated using 12 compounds and predicted 11 out of 12 of the Ki-values within 0.5 log units. The discriminant model will be useful in screening for CYP2C9 inhibitors from large compound collections. The 3D-QSAR model will be used during lead optimization to avoid chemistry that result in inhibition of CYP2C9.
The recently introduced GRid-INdependent Descriptors (GRIND) were designed to provide a suitable description of a series of ligands for 3D-QSAR studies not requiring the spatial superimposition of their structures. Despite the proven usefulness of the method, it was recognized that the original GRIND failed to describe appropriately the shape of the ligand molecules, which in some cases plays a major role in ligand-receptor binding. For this reason, the original descriptors have been enhanced with the addition of a molecular shape description based on the local curvature of the molecular surface. The integration of this description into the GRIND allows the generation of 3D-QSAR models able to identify both favorable and unfavorable shape complementarity in a simple and alignment-independent way. The usefulness of the new GRIND-shape description in 3D-QSAR is illustrated using two structure-activity studies: one performed on a set of xanthine-like antagonists of the A(1) adenosine receptor; another performed on a series of Plasmodium falciparum plasmepsin II inhibitors.
We have screened the entire Protein Data Bank (Release No. 103, January 2003) and identified 5671 protein-ligand complexes out of 19 621 experimental structures. A systematic examination of the primary references of these entries has led to a collection of binding affinity data (K(d), K(i), and IC(50)) for a total of 1359 complexes. The outcomes of this project have been organized into a Web-accessible database named the PDBbind database.
A powerful in silico strategy based on the combined use of two computational tools (MLP and MIFs) able to calculate and visualize 3D molecular fields can give useful information about surface properties of macromolecules involved in the mechanisms of formation of complexes. In particular, this study investigated the variation in polar/hydrophobic pattern induced on the beta-CD alone (i.e. =without the ligand) by the inclusion of four ligands having different lipophilicities and small size. Results indicate that, in the presence of guests with P>0, the hydrophobicity of beta-CD increases in the cavity and its surroundings on the primary face.
The paper describes the generation of four types of three-dimensional molecular field descriptors or 'field points' as extrema of electrostatic, steric, and hydrophobic fields. These field points are used to define the properties necessary for a molecule to bind in a characteristic way into a specified active site. The hypothesis is that compounds showing a similar field point pattern are likely to bind at the same target site regardless of structure. The methodology to test this idea is illustrated using HIV NNRTI and thrombin ligands and validated across seven other targets. From the in silico comparisons of field point overlays, the experimentally observed binding poses of these ligands in their respective sites can be reproduced from pairwise comparisons.
Mapping interactions at protein-ligand binding sites is an important aspect of understanding many biological reactions and a key part of drug design. In this paper, we have used a fragment-based approach to probe "hot spots" at the cofactor-binding site of a model dehydrogenase, Escherichia coli ketopantoate reductase. Our strategy involved the breaking down of NADPH (Kd = 300 nM) into smaller fragments and the biophysical characterization of their binding using WaterLOGSY NMR spectroscopy, isothermal titration calorimetry (ITC), and inhibition studies. The weak binding affinities of fragments were measured by direct ITC titrations under low c value conditions. The 2'-phosphate and the reduced nicotinamide groups were found to contribute a large part of the binding energy. A combination of ITC and site-directed mutagenesis enabled us to locate the fragments at separate hot spots on opposite ends of the cofactor-binding site. This study has identified structural determinants for cofactor recognition that represent a blueprint for future inhibitor design.
) GRID v22. Molecular Discovery Ltd
(8) GRID v22. Molecular Discovery Ltd. London. UK. 2006.