Docking and pharmacophore‐based alignment comparative molecular field analysis three‐dimensional quantitative structure–activity relationship analysis of dihydrofolate reductase inhibitors by linear and nonlinear calibration methods

Ghasemi, Jahan B.; Meftahi, Nastaran; Pirhadi, Somayeh; Tavakoli, Hossein

doi:10.1002/cem.2515

“…Finally, we used machine learning methods to explore the contributions of the parameters (model type, partial charge, Lennard-Jones parameters, bond length and angle) to the surface tension, dielectric constant, and self-diffusion coefficient. Machine learning can be used to predict a wide variety of material, [86][87][88][89] chemical, [90][91][92][93] and biological properties, [94][95][96][97] with several commercial and non-commercial open-source platforms that can be used to develop machine learning algorithms such as Schrödinger, 98 SYBYL, 99 TensorFlow (Google), 100 and BioPPSy. 101…”

Section: Name Typementioning

confidence: 99%

A systematic comparison of the structural and dynamic properties of commonly used water models for molecular dynamics simulations

Pathirannahalage

¹

,

Meftahi

²

,

Elbourne

³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Water is a unique solvent that is ubiquitous in biology and present in a variety of solutions, mixtures, and materials settings. It therefore forms the basis for all molecular dynamics simulations of biological phenomena, as well as for many chemical, industrial, and materials investigations. Over the years, many water models have been developed, and it remains a challenge to find a single water model that accurately reproduces all experimental properties of water simultaneously. Here, we report a comprehensive comparison of structural and dynamic properties of 30 commonly used 3-point, 4-point, 5-point, and polarizable water models simulated using consistent settings and analysis methods. For the properties of density, coordination number, surface tension, dielectric constant, self-diffusion coefficient, and solvation free energy of methane, models published within the past two decades consistently show better agreement with experimental values compared to models published earlier, albeit with some notable exceptions. However, no single model reproduced all experimental values exactly, highlighting the need to carefully choose a water model for a particular study, depending on the phenomena of interest. Finally, machine learning algorithms quantified the relationship between the water model force field parameters and the resulting bulk properties, providing insight into the parameter-property relationship and illustrating the challenges of developing a water model that can accurately reproduce all properties of water simultaneously.

show abstract

“…The proper α is estimated by cross‐validation to optimize the right size of the tree as a balance between complexity and performance .…”

Section: Theorymentioning

confidence: 99%

“…In the absence of experimental data, a quantitative structure activity relationship (QSAR) is a good remedy for the prediction of activity in the biological systems . Generally, the ability of multivariate regression methods in mapping a chemical space to the relevant biological activity landscape is crucial in complex nonlinear biological systems.…”

Section: Introductionmentioning

confidence: 99%

An improved ensemble learning machine for biological activity prediction of tyrosine kinase inhibitors

Tavakoli

¹

,

Ghasemi

²

2015

Journal of Chemometrics

Self Cite

View full text Add to dashboard Cite

Boosting is one of the most important strategies in ensemble learning because of its ability to improve the stability and performance of weak learners. It is nonparametric, multivariate, fast and interpretable but is not robust against outliers. To enhance its prediction accuracy as well as immunize it against outliers, a modified version of a boosting algorithm (AdaBoost R2) was developed and called AdaBoost R3. In the sampling step, extremum samples were added to the boosting set. In the robustness step, a modified Huber loss function was applied to overcome the outlier problem. In the output step, a deterministic threshold was used to guarantee that bad predictions do not participate in the final output. The performance of the modified algorithm was investigated with two anticancer data sets of tyrosine kinase inhibitors, and the mechanism of inhibition was studied using the relative weighted variable importance procedure. Investigating the effect of base learner's strength reveals that boosting is only successful using the classification and regression tree method (a weak to moderate learner) and does not have a significant effect using the radial basis functions partial least square method (a strong base learners).

show abstract

“…85 20 Finally, we used machine learning methods to explore the contributions of the parameters (model type, partial charge, Lennard-Jones parameters, bond length and angle) to the surface tension, dielectric constant, and self-diffusion coefficient. Machine learning can be used to predict a wide variety of material, [86][87][88][89] chemical, [90][91][92][93] and biological properties, [94][95][96][97] with several commercial and non-commercial open-source platforms that can be used to develop machine learning algorithms such as Schrödinger, 98 SYBYL, TensorFlow (Google), 100 and BioPPSy.…”

Section: Introductionmentioning

confidence: 99%

A systematic comparison of the structural and dynamic properties of commonly used water models for molecular dynamics simulations

Pathirannahalage

¹

,

Meftahi

²

,

Elbourne

³

et al. 2021

Preprint

Self Cite

3

0

View full text Add to dashboard Cite

Water is a unique solvent that is ubiquitous in biology and present in a variety of solutions, mixtures, and materials settings. It therefore forms the basis for all molecular dynamics simulations of biological phenomena, as well as for many chemical, industrial, and materials investigations. Over the years, many water models have been developed, and it remains a challenge to find a single water model that accurately reproduces all experimental properties of water simultaneously. Here, we report a comprehensive comparison of structural and dynamic properties of 30 commonly used 3-point, 4-point, 5-point, and polarizable water models simulated using consistent settings and analysis methods. For the properties of density, coordination number, surface tension, dielectric constant, self-diffusion coefficient, and solvation free energy of methane, models published within the past two decades consistently show better agreement with experimental values compared to models published earlier, albeit with some notable exceptions. However, no single model reproduced all experimental values exactly, highlighting the need to carefully choose a water model for a particular study, depending on the phenomena of interest. Finally, machine learning algorithms quantified the relationship between the water model force field parameters and the resulting bulk properties, providing insight into the parameter-property relationship and illustrating the challenges of developing a water model that can accurately reproduce all properties of water simultaneously. referred to as TIPS3P hereafter in this work) which is used with the Chemistry at HARvard Macromolecular Mechanics (CHARMM) force field. 46 In the same 1983 paper as the original TIP3P model, Jorgensen et al. also described a 4-point model, TIP4P, where the negative charge of the oxygen was offset from the Lennard-Jones potential, resulting in improved structural properties. 41 The TIP4P water model is also commonly used with the OPLS force field. For the Consistent-Valence Force Field (CVFF), the CVFF water model has a geometry similar to TIP3P and non-bond parameters similar to SPC. It was developed concurrently with the rest of the force field, and is notable for having a flexible geometry, the use of a Morse potential for the O-H bond, and cross-terms to reproduce vibrational frequencies. 47 These models remain a popular choice due to their simple geometry, low computational cost, and relative accuracy. With the exception of CVFF, these models have a rigid geometry, which enables the use of a larger (e.g. 2 fs) timestep in MD simulations. While these models reproduced some experimental properties of water reasonably well, limited computational resources available at the time necessitated small system sizes of 125 to 216 water molecules, short simulation times of less than 20 ps, and short cut-offs of non-bond interactions typically between 7.5 and 8.5 Å. Moreover, many structural properties, such as densities and dielectric constants, were calculated using Monte Carlo, rather tha...

show abstract

Docking and pharmacophore‐based alignment comparative molecular field analysis three‐dimensional quantitative structure–activity relationship analysis of dihydrofolate reductase inhibitors by linear and nonlinear calibration methods

Cited by 14 publications

References 38 publications

A systematic comparison of the structural and dynamic properties of commonly used water models for molecular dynamics simulations

A systematic comparison of the structural and dynamic properties of commonly used water models for molecular dynamics simulations

An improved ensemble learning machine for biological activity prediction of tyrosine kinase inhibitors

A systematic comparison of the structural and dynamic properties of commonly used water models for molecular dynamics simulations

Contact Info

Product

Resources

About