Nowadays, in silico methodologies have become a crucial part of the drug discovery process. This is mostly because they can impact the entire drug development trajectory, identifying and discovering new potential drugs with a significant reduction to cost and time. Furthermore, computer-aided drug design (CADD) approaches are important for reducing the experimental use of animals for in vivo testing, for aiding the design of safer drugs, and for repositioning known drugs, assisting medicinal chemists at each step (design, discovery, development, and hit-optimization) during the drug discovery process. On one hand, conventional methods for drug discovery involve the costly random screening of synthesized compounds or natural products. On the other hand, computational procedures can be very multifarious, requiring interdisciplinary studies and the application of computer science to rationally design effective and commercially feasible drugs.
Remarkable progress has been made both in computer science (accelerating drug discovery research), and the development of new experimental procedures for the characterization of biological targets. Among methods employed in drug discovery, pharmacophore modelling, multi-dimensional quantitative structure activity relationships (such as, 4- and 3D-QSAR), Comparative Molecular Field Analysis (CoMFA), and Comparative Molecular Similarity Indices Analysis (CoMSIA) remain the preferred ligand-based (LB) methods for fast virtual screening (VS) procedures and for rationalizing the activities of a set of ligands. In a recent breakthrough, a novel approach in the QSAR field has been represented by the combination of Molecular Dynamics (MD), relative computed descriptors, and the generation of QSAR models. This approach provides computational tools, the so-called MD-QSAR models, with enhanced predictive power. Once information about the 3D structure of the targets in complex with ligands has been elucidated, structure-based (SB) drug design approaches like SB pharmacophore models—which include excluded volumes or high throughput dockings—are the elected methods for identifying novel chemical entities for a selected target. At the same time, if we wish to investigate ligand-receptor complexes and in general the dynamics and thermodynamics of biological systems, MD simulations represent one of the main computational resources and still remain the most representative technique for this kind of investigation. When the structure of a target protein is unknown, with only the amino acids sequence available, homology modelling techniques can be employed to build the 3D structure.
In addition, to better characterize biological systems, thus understanding the mechanism of action of enzymes that are in complexes with ligands, quantum mechanics/molecular mechanics (QM/MM) calculations can be helpful. Currently, QM/MM can be combined with MD (QM/MM-MD) to completely characterize enzymatic mechanisms. As many configurations are generated in a sufficiently large MD, the number of QM calculations required in a QM/MM–MD simulation can become too high. Aiming, then, to reduce the computational burden, without loss of information, modern theoretical methods for selection of relevant configurations can be employed. These tools can help the scientists to shorten the cycle of drug discovery, and thus make the process more cost-affordable. Significant technological gains in hardware and software resources, algorithm design, as well as biological advances in identifying new drug targets, have made computer-assisted approaches the most valuable methods in pre-clinical research.
As such, this Research Topic welcomes submissions from researchers in the field of computational drug discovery and design, including original research and review articles related to the in silico approaches used in Medicinal Chemistry. Please note that computational studies must demonstrate a clear use in medicinal chemistry through comparison with experimental data and experimental validation.
Nowadays, in silico methodologies have become a crucial part of the drug discovery process. This is mostly because they can impact the entire drug development trajectory, identifying and discovering new potential drugs with a significant reduction to cost and time. Furthermore, computer-aided drug design (CADD) approaches are important for reducing the experimental use of animals for in vivo testing, for aiding the design of safer drugs, and for repositioning known drugs, assisting medicinal chemists at each step (design, discovery, development, and hit-optimization) during the drug discovery process. On one hand, conventional methods for drug discovery involve the costly random screening of synthesized compounds or natural products. On the other hand, computational procedures can be very multifarious, requiring interdisciplinary studies and the application of computer science to rationally design effective and commercially feasible drugs.
Remarkable progress has been made both in computer science (accelerating drug discovery research), and the development of new experimental procedures for the characterization of biological targets. Among methods employed in drug discovery, pharmacophore modelling, multi-dimensional quantitative structure activity relationships (such as, 4- and 3D-QSAR), Comparative Molecular Field Analysis (CoMFA), and Comparative Molecular Similarity Indices Analysis (CoMSIA) remain the preferred ligand-based (LB) methods for fast virtual screening (VS) procedures and for rationalizing the activities of a set of ligands. In a recent breakthrough, a novel approach in the QSAR field has been represented by the combination of Molecular Dynamics (MD), relative computed descriptors, and the generation of QSAR models. This approach provides computational tools, the so-called MD-QSAR models, with enhanced predictive power. Once information about the 3D structure of the targets in complex with ligands has been elucidated, structure-based (SB) drug design approaches like SB pharmacophore models—which include excluded volumes or high throughput dockings—are the elected methods for identifying novel chemical entities for a selected target. At the same time, if we wish to investigate ligand-receptor complexes and in general the dynamics and thermodynamics of biological systems, MD simulations represent one of the main computational resources and still remain the most representative technique for this kind of investigation. When the structure of a target protein is unknown, with only the amino acids sequence available, homology modelling techniques can be employed to build the 3D structure.
In addition, to better characterize biological systems, thus understanding the mechanism of action of enzymes that are in complexes with ligands, quantum mechanics/molecular mechanics (QM/MM) calculations can be helpful. Currently, QM/MM can be combined with MD (QM/MM-MD) to completely characterize enzymatic mechanisms. As many configurations are generated in a sufficiently large MD, the number of QM calculations required in a QM/MM–MD simulation can become too high. Aiming, then, to reduce the computational burden, without loss of information, modern theoretical methods for selection of relevant configurations can be employed. These tools can help the scientists to shorten the cycle of drug discovery, and thus make the process more cost-affordable. Significant technological gains in hardware and software resources, algorithm design, as well as biological advances in identifying new drug targets, have made computer-assisted approaches the most valuable methods in pre-clinical research.
As such, this Research Topic welcomes submissions from researchers in the field of computational drug discovery and design, including original research and review articles related to the in silico approaches used in Medicinal Chemistry. Please note that computational studies must demonstrate a clear use in medicinal chemistry through comparison with experimental data and experimental validation.
