A combined Fisher and Laplacian score for feature selection in QSAR based drug design using compounds with known and unknown activities

Hasanloei, Mohammad Amin Valizade; Sheikhpour, Razieh; Sarram, Mehdi Agha; Sheikhpour, Elnaz; Sharifi, Hamdollah

doi:10.1007/s10822-017-0094-6

Cited by 18 publications

(3 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In-house scripts using the BINANA algorithm (software used in other non-GPCR studies [39,75,76,77,78]) were constructed to identify the type of interactions established between the ligands and binding pocket amino-acids [39]. We measured close contacts between receptor and ligands below or equal 2.5 Å and below or equal 4.0 Å, hydrogen bonds (HB), hydrophobic contacts (hydrocontacts) and salt-bridges (SB) as well as π-interactions, further subdivided into cation-π-interactions (cat-π), aromatic superpositions (π-π-stack) and perpendicular interactions of aromatic rings also referred to as edge-face-interactions (T-stack) [39].…”

Section: Resultsmentioning

confidence: 99%

A Complete Assessment of Dopamine Receptor- Ligand Interactions through Computational Methods

Bueschbell

Barreto²,

Preto³

et al. 2019

Molecules

View full text Add to dashboard Cite

Background: Selectively targeting dopamine receptors (DRs) has been a persistent challenge in the last years for the development of new treatments to combat the large variety of diseases involving these receptors. Although, several drugs have been successfully brought to market, the subtype-specific binding mode on a molecular basis has not been fully elucidated. Methods: Homology modeling and molecular dynamics were applied to construct robust conformational models of all dopamine receptor subtypes (D1-like and D2-like). Fifteen structurally diverse ligands were docked. Contacts at the binding pocket were fully described in order to reveal new structural findings responsible for selective binding to DR subtypes. Results: Residues of the aromatic microdomain were shown to be responsible for the majority of ligand interactions established to all DRs. Hydrophobic contacts involved a huge network of conserved and non-conserved residues between three transmembrane domains (TMs), TM2-TM3-TM7. Hydrogen bonds were mostly mediated by the serine microdomain. TM1 and TM2 residues were main contributors for the coupling of large ligands. Some amino acid groups form electrostatic interactions of particular importance for D1R-like selective ligands binding. Conclusions: This in silico approach was successful in showing known receptor-ligand interactions as well as in determining unique combinations of interactions, which will support mutagenesis studies to improve the design of subtype-specific ligands.

show abstract

Section: Resultsmentioning

confidence: 99%

A Complete Assessment of Dopamine Receptor- Ligand Interactions through Computational Methods

Bueschbell

Barreto²,

Preto³

et al. 2019

Molecules

View full text Add to dashboard Cite

show abstract

“…BINANA (used in other non-GPCR studies [30,[67][68][69][70]) revealed to be a helpful tool for assessing the full binding capacity of the DRs regarding the chosen ligand set. First of all, it was visible by considering the total number of interactions per ligand that no clear D1-or D2-like specificity was observed, except for apomorphine (differences of 20 interactions between D1-like and D2-like).…”

Section: Pairwise Interactionsmentioning

confidence: 99%

A Complete Assessment of Dopamine Receptor-Ligand Interactions through Computational Methods

Bueschbell¹,

Barreto²,

Preto³

et al. 2019

Preprint

View full text Add to dashboard Cite

Background: Selectively targeting dopamine receptors has been a persistent challenge in the last years for the development of new treatments to combat the large variety of diseases evolving these receptors. Although, several drugs have been successfully brought to market, the subtype-specific binding mode on a molecular basis has not been fully elucidated.Methods: Homology modeling and molecular dynamics were applied to construct robust conformational models of all dopamine receptor subtypes (D1-like and D2like receptors). Fifteen structurally diverse ligands were docked to these models. Contacts at the binding pocket were fully described in order to reveal new structural findings responsible for DR sub-type specificity. Results:We showed that the number of conformations for a receptor:ligand complex was associated to unspecific interactions > 2.5 Å and hydrophobic contacts, while the decoys binding energy was influenced by specific electrostatic interactions. Known residues such as 3.32Asp, the serine microdomain and the aromatic microdomain were found interacting in a variety of modes (HB, SB, πstacking). Purposed TM2-TM3-TM7 microdomain was found to form a hydrophobic network involving Orthosteric Binding Pocket (OBP) and Secondary Binding Pocket (SBP). T-stacking interactions revealed as especially relevant for some large ligands such as apomorphine, risperidone or aripiprazole.Conclusions: This in silico approach was successful in showing known receptorligand interactions as well as in determining unique combinations of interactions, key for the design of more specific ligands.The strive for finding new and effective therapeutics led to a growing interest in the use of Computer Aided Drug Design (CADD). Originally developed for High-Throughput Screening (HTS) of compound libraries, the use of CADD nowadays plays an important role in drug discovery [24]. Modeling three-dimensional (3D) target proteins help to visualize, analyse and optimize known ligands and discover new lead compounds [25]. The CADD pipeline can be classified in two general categories: structure-based and ligand-based, dependent on the available information about the topic of investigation [25]. A structure-based CADD is used when the target, e.g. a receptor, is known and so compound libraries can be screened to find suitable structures for the target. Usually protein-ligand docking studies are performed or ligands are designed de novo and are then used for compound library screening to test possible lead structures experimentally. Vice-versa, a ligand-based CAAD procedure is used when ligand structure information is provided to create pharmacophore models and to perform virtual screening [24]. All in all, CADD faces the challenges of identifying novel targets and their ligands, for example to treat common and rare diseases [26]. AimModeling G protein-coupled receptors (GPCRs) remains challenging due to the complex structure of these membrane proteins and the lack of structural information about the desired receptor to target, however CADD m...

show abstract

“…The structure-based molecular design mainly includes a receptor-based method through a three-dimensional (3D) chemical structure to obtain ligand interaction [1,35,36]. However, traditional QSAR models may frequently miss suitable candidate molecules, because of the poor predictive accuracy and versatility caused by poor feature selection that requires skill and knowledge and conformational limitations for coincidence effect [1,[37][38][39]. Therefore, a QSAR system with high-throughput and performance is desired because of the development of novel medicines, chemicals, and nanomaterials on human health.…”

Section: Introductionmentioning

confidence: 99%

A Deep Learning-Based Quantitative Structure–Activity Relationship System Construct Prediction Model of Agonist and Antagonist with High Performance

Matsuzaka

Uesawa

2022

IJMS

View full text Add to dashboard Cite

Molecular design and evaluation for drug development and chemical safety assessment have been advanced by quantitative structure–activity relationship (QSAR) using artificial intelligence techniques, such as deep learning (DL). Previously, we have reported the high performance of prediction models molecular initiation events (MIEs) on the adverse toxicological outcome using a DL-based QSAR method, called DeepSnap-DL. This method can extract feature values from images generated on a three-dimensional (3D)-chemical structure as a novel QSAR analytical system. However, there is room for improvement of this system’s time-consumption. Therefore, in this study, we constructed an improved DeepSnap-DL system by combining the processes of generating an image from a 3D-chemical structure, DL using the image as input data, and statistical calculation of prediction-performance. Consequently, we obtained that the three prediction models of agonists or antagonists of MIEs achieved high prediction-performance by optimizing the parameters of DeepSnap, such as the angle used in the depiction of the image of a 3D-chemical structure, data-split, and hyperparameters in DL. The improved DeepSnap-DL system will be a powerful tool for computer-aided molecular design as a novel QSAR system.

show abstract

A combined Fisher and Laplacian score for feature selection in QSAR based drug design using compounds with known and unknown activities

Cited by 18 publications

References 32 publications

A Complete Assessment of Dopamine Receptor- Ligand Interactions through Computational Methods

A Complete Assessment of Dopamine Receptor- Ligand Interactions through Computational Methods

A Complete Assessment of Dopamine Receptor-Ligand Interactions through Computational Methods

A Deep Learning-Based Quantitative Structure–Activity Relationship System Construct Prediction Model of Agonist and Antagonist with High Performance

Contact Info

Product

Resources

About