Molecular docking methodologies

Directional ΔG Neural Network (DrΔG-Net): A Modular Neural Network Approach to Binding Free Energy Prediction

The protein-ligand binding free energy is a central quantity in structure-based computational drug discovery efforts. Although popular alchemical methods provide sound statistical means of computing the binding free energy of a large breadth of systems, they are generally too costly to be applied at the same frequency as end point or ligand-based methods. By contrast, these data-driven approaches are typically fast enough to address thousands of systems but with reduced transferability to unseen systems. We introduce DrΔG-Net (or simply Dragnet), an equivariant graph neural network that can blend ligand-based and protein-ligand data-driven approaches. It is based on a 3D fingerprint representation of the ligand alone and in complex with the protein target. Dragnet is a global scoring function to predict the binding affinity of arbitrary protein-ligand complexes, but can be easily tuned via transfer learning to specific systems or end points, performing similarly to common 2D ligand-based approaches in these tasks. Dragnet is evaluated on a total of 28 validation proteins with a set of congeneric ligands derived from the Binding DB and one custom set extracted from the ChEMBL Database. In general, a handful of experimental binding affinities are sufficient to optimize the scoring function for a particular protein and ligand scaffold. When not available, predictions from physics-based methods such as absolute free energy perturbation can be used for the transfer learning tuning of Dragnet. Furthermore, we use our data to illustrate the present limitations of data-driven modeling of binding free energy predictions.

Molecular docking-based interaction studies on imidazo[1,2-a] pyridine ethers and squaramides as anti-tubercular agents

Development of new anti-tubercular agents is required in the wake of resistance to the existing and newly approved drugs through novel-validated targets like ATP synthase, etc. The major limitation of poor correlation between docking scores and biological activity by SBDD was overcome by a novel approach of quantitatively correlating the interactions of different amino acid residues present in the target protein structure with the activity. This approach well predicted the ATP synthase inhibitory activity of imidazo[1,2-a] pyridine ethers and squaramides (r = 0.84) in terms of Glu65b interactions. Hence, the models were developed on combined (r = 0.78), and training (r = 0.82) sets of 52, and 27 molecules, respectively. The training set model well predicted the diverse dataset (r = 0.84), test set (r = 0.755), and, external dataset (r_ext = 0.76). This model predicted three compounds from a focused library generated by incorporating the essential features of the ATP synthase inhibition with the pIC₅₀ values in the range of 0.0508-0.1494 µM. Molecular dynamics simulation studies ascertain the stability of the protein structure and the docked poses of the ligands. The developed model(s) may be useful in the identification and optimization of novel compounds against TB.

A new insight of immunosuppressive microenvironment in osteosarcoma lung metastasis

The lung is the primary organ for the metastasis of osteosarcoma. Although the application of neoadjuvant chemotherapy and surgery has remarkably improved the survival rate of patients with osteosarcoma, prognosis is still poor for those patients with metastasis. In this study, we performed further bioinformatics analysis on single-cell RNA sequencing (scRNA-seq) data published before, containing 75,317 cells from two osteosarcoma lung metastasis and five normal lung tissues. First, we classified 17 clusters, including macrophages, T cells, endothelial cells, and so on, indicating highly intratumoral heterogeneity in osteosarcoma lung metastasis. Next, we found macrophages in osteosarcoma lung metastasis did not have significant M1 or M2 polarizations. Then, we identified that T cells occupied the most abundant among all cell clusters, and found CD8+ T cells exhibited a low expression level of immune checkpoints in osteosarcoma lung metastasis. What is more, we identified C2_Malignant cells, and found CD63 might play vital roles in determining the infiltration of T cells and malignant cells in conventional-type osteosarcoma lung metastasis. Finally, we unveiled C1_Therapeutic cluster, a subcluster of malignant cells, was sensitive to oxfendazole and mevastatin, and the potential hydrogen-bond position and binding energy of oxfendazole-KIAA0907 and mevastatin-KIAA0907 were unveiled, respectively. Our results highlighted the power of scRNA-seq technique in identifying the complex tumor microenvironment of osteosarcoma lung metastasis, making it possible to devise precision therapeutic approaches.

New Indole Diketopiperazine Alkaloids from Soft Coral-Associated Epiphytic Fungus Aspergillus versicolor CGF 9-1-2

Two new indole diketopiperazine alkaloids (IDAs), (+)19-epi-sclerotiamide (1) and (-)19-epi-sclerotiamide (2), along with 13 known analogs (3-15), were isolated from a soft coral-associated epiphytic fungus Aspergillus versicolor CGF 9-1-2. The structures of two new compounds were established based on the combination of HR-ESI-MS, 1D and 2D NMR spectroscopy, optical rotation measurements and quantum chemical 13 C-NMR, the absolute configurations were determined by experimental and electronic circular dichroism (ECD) calculations. The results of molecular docking showed that all the compounds had a good binding with TDP1, TDP2, TOP1, TOP2, Ache, NLRP3, EGFR, EGFR L858R, EGFR T790M and EGFR T790/L858. Biological evaluation of compounds 3, 6, 8, 11 showed that 3 exerted a strong inhibitory effect on TDP2 with a rate of 81.72 %.

Molecular docking-based interactions in QSAR studies on Mycobacterium tuberculosis ATP synthase inhibitors

Tuberculosis (TB) is a global threat with a large burden across the continents in terms of mortality, morbidity, and financial losses. The disease has evolved into multi-drug-resistant (MDR-TB) and extensively drug-resistant (XDR-TB) tuberculosis owing to numerous factors ranging from patients' non-compliance to demographical implications. There have been very few new drugs for resistant TB. Resistance has already been reported even for the newly introduced drug bedaquiline. An attempt has been made to integrate both structure-based and QSAR drug design techniques (QSAR-SBDD) for the identification of novel leads. The docking scores normally do not correlate with the activity. Hence, the docking results have been analysed in terms of the number of interactions rather than docking scores. The parameters derived from interactions have been used in developing the QSAR models. The best model shows a good correlation (r = 0.908) between the activity and interaction parameter 'C' describing the sum of all the interactions with each amino acid residue. This model also predicts external dataset with a good correlation (r_ext = 0.851) and can be used for the identification of novel chemical entities (NCEs) and repurposed drugs for TB therapeutics.

Application of Machine Learning Techniques in Drug-target Interactions Prediction

BACKGROUND

Drug-Target interactions are vital for drug design and drug repositioning. However, traditional lab experiments are both expensive and time-consuming. Various computational methods which applied machine learning techniques performed efficiently and effectively in the field.

RESULTS

The machine learning methods can be divided into three categories basically: Supervised methods, Semi-Supervised methods and Unsupervised methods. We reviewed recent representative methods applying machine learning techniques of each category in DTIs and summarized a brief list of databases frequently used in drug discovery. In addition, we compared the advantages and limitations of these methods in each category.

CONCLUSION

Every prediction model has both strengths and weaknesses and should be adopted in proper ways. Three major problems in DTIs prediction including the lack of nonreactive drug-target pairs data sets, over optimistic results due to the biases and the exploiting of regression models on DTIs prediction should be seriously considered.

Inhibition of Acetylcholinesterase by Wood Creosote and Simple Phenolic Compounds

Anisakiasis is common in countries where raw or incompletely cooked marine fish are consumed. Currently, effective therapeutic methods to treat anisakiasis are unavailable. A recent study found that wood creosote inactivates the movement of Anisakis species. Essential oil of Origanum compactum containing carvacrol and thymol, which are similar to the constituents of wood creosote, was reported to inactivate Anisakis by inhibiting its acetylcholinesterase. We examined whether wood creosote can also inhibit acetylcholinesterase. We examined the effect of components of wood creosote using the same experimental method. A computer simulation experiment (molecular docking) was also performed. Here, we demonstrate that wood creosote inactivated acetylcholinesterase in a dose-dependent manner with an IC₅₀ of 0.25 mg/mL. Components of wood creosote were also tested individually: 5-methylguaiacol, p-cresol, guaiacol, o-cresol, 2,4-dimethylphenol, m-cresol, phenol and 4-methylguaiacol inactivated the enzyme with an IC₅₀ of 14.0, 5.6, 17.0, 6.3, 3.9, 10.0, 15.2 and 27.2 mM, respectively. The mechanism of acetylcholinesterase inactivation was analyzed using a computer-based molecular docking simulation, which employed a three-dimensional structure of acetylcholinesterase and above phenolic compounds as docking ligands. The simulation indicated that the phenolic compounds bind to the active site of the enzyme, thereby competitively blocking entry of the substrate acetylcholine. These findings suggest that the mechanism for the inactivation of Anisakis movement by wood creosote is due to inhibition of acetylcholinesterase needed for motor neuron activity.

In Silico 3D Modeling of Binding Activities

In silico three-dimensional (3D) molecular modeling tools based upon the receptor/enzyme-ligand docking simulation in protein crystal structures and/or homology modeling of receptors have been reliably used in pharmacological research and development for decades. Molecular docking methodologies are helpful for revealing facets of activation and inactivation, thus improving mechanistic understanding and predicting molecular ligand binding activity, and they can have a high level of accuracy, and have also been explored and applied in chemical risk assessment. This computational approach is, however, only applicable for chemical hazard identification situations where the specific target receptor for a given chemical is known and the crystal structure/homology model of the receptor is available.

Current Trends in Docking Methodologies

Molecular docking was earlier considered to predict the binding affinity of the receptor and ligand molecules. With the progress in computational power and developing approaches, new horizons are now opening for accurate prediction of molecular binding affinity. In the current book chapter, recent strategies for Computer-Aided Drug Designing (CADD) including virtual screening and molecular docking, encompassing molecular dynamics simulations, and binding free energy calculation methods are discussed. Brief overview of different binding free energy methods MMPBSA, MMGBSA, LIE and TI have also been given along with the recent Relaxed Complex Scheme protocol.

Structure-activity relationships of FMRF-NH2 peptides demonstrate A role for the conserved C terminus and unique N-terminal extension in modulating cardiac contractility

FMRF-NH2 peptides which contain a conserved, identical C-terminal tetrapeptide but unique N terminus modulate cardiac contractility; yet, little is known about the mechanisms involved in signaling. Here, the structure-activity relationships (SARs) of the Drosophila melanogaster FMRF-NH2 peptides, PDNFMRF-NH2, SDNFMRF-NH2, DPKQDFMRF-NH2, SPKQDFMRF-NH2, and TPAEDFMRF-NH2, which bind FMRFa-R, were investigated. The hypothesis tested was the C-terminal tetrapeptide FMRF-NH2, particularly F1, makes extensive, strong ligand-receptor contacts, yet the unique N terminus influences docking and activity. To test this hypothesis, docking, binding, and bioactivity of the C-terminal tetrapeptide and analogs, and the FMRF-NH2 peptides were compared. Results for FMRF-NH2 and analogs were consistent with the hypothesis; F1 made extensive, strong ligand-receptor contacts with FMRFa-R; Y → F (YMRF-NH2) retained binding, yet A → F (AMRF-NH2) did not. These findings reflected amino acid physicochemical properties; the bulky, aromatic residues F and Y formed strong pi-stacking and hydrophobic contacts to anchor the ligand, interactions which could not be maintained in diversity or number by the small, aliphatic A. The FMRF-NH2 peptides modulated heart rate in larva, pupa, and adult distinctly, representative of the contact sites influenced by their unique N-terminal structures. Based on physicochemical properties, the peptides each docked to FMRFa-R with one best pose, except FMRF-NH2 which docked with two equally favorable poses, consistent with the N terminus influencing docking to define specific ligand-receptor contacts. Furthermore, SDNAMRF-NH2 was designed and, despite lacking the aromatic properties of one F, it binds FMRFa-R and demonstrated a unique SAR, consistent with the N terminus influencing docking and conferring binding and activity; thus, supporting our hypothesis.