Multiple field three dimensional quantitative structure–activity relationship (MF‐3D‐QSAR)

Du, Qiang; Huang, Ribo; Wei, Yutuo; Du, Liqin; Chou, Kuo‐Chen

doi:10.1002/jcc.20776

Cited by 90 publications

(56 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional QSAR models have recently been reconstructed to include more sophisticated theoretically based approaches and mathematical tools. More sophisticated QSAR approaches such as fragment-based two dimensional QSAR models and multiple field three-dimensional QSAR models have been recently developed [17,18]. These methods remarkably enhance the predictive power of QSAR models and additionally provide one with more molecular structural information than conventional or traditional QSAR.…”

Section: Introductionmentioning

confidence: 99%

A Machine Learning Approach to Designing Guidelines for Acute Aquatic Toxicity

Husowitz¹,

Sanchez-Arias²

2017

J Biom Biostat

View full text Add to dashboard Cite

A support vector classification wrapper feature elimination approach was used to find the most relevant pairs of molecular features that adequately and accurately can predict acute aquatic toxicity. These pairs were then used to derive chemical thresholds or boundaries between chemical properties for toxic and nontoxic organic chemicals that can be used as a "rule of thumb" to design less toxic chemicals. The most relevant pairs were determined to be: Lowest Unoccupied Molecular Orbital (LUMO) and Aqueous Solubility (QPlogS), Difference between the LUMO and HOMO (dE) and Octonal-Water Partition Coefficient (QPlogo.w), and Difference between the LUMO and HOMO (dE) and Van der Waals surface area of polar nitrogen and oxygen atoms (PSA). Projected hyper planes were constructed for each pair and the following thresholds were found: for Lowest Unoccupied Molecular Orbital (LUMO) and Aqueous Solubility (QPlogS) they roughly correspond to QPlogS>-1 and LUMO>1, and for Octonal-Water Partition Coefficient (QPlogo.w) vs. difference between the LUMO and HOMO (dE) they roughly correspond to QPlogo.w<1 and dE>9. This study shows how a statistical approach such as support vector machines can be applied to the rational design of chemicals with reduced toxicity.Citation: Husowitz B, Sanchez-Arias R (2017) A Machine Learning Approach to Designing Guidelines for Acute Aquatic Toxicity. J Biom Biostat 8:385.

show abstract

Section: Introductionmentioning

confidence: 99%

A Machine Learning Approach to Designing Guidelines for Acute Aquatic Toxicity

Husowitz¹,

Sanchez-Arias²

2017

J Biom Biostat

View full text Add to dashboard Cite

show abstract

“…It is a combination and advanced form of classical 2D-QSAR and traditional 3D-QSAR. In addition to the electrostatic and van der Waals potentials, more potential fields (such as lipophilic potential, hydrogen bonding potential, and non-thermodynamic factors) are integrated in the MF-3D-QSAR [9].…”

Section: Introductionmentioning

confidence: 99%

QSAR and its Role in Target-Ligand Interaction

Singh¹,

Singh²

2013

TOBIOIJ

View full text Add to dashboard Cite

Each molecule has its own specialty, structure and function and when these molecules are combined together they form a compound. Structure and function of a molecule are related to each other and QSARs (Quantitative StructureActivity relationships) are based on the criteria that the structure of a molecule must contain the features responsible for its physical, chemical, and biological properties, and on the ability to represent the chemical by one, or more, numerical descriptor(s). By QSAR models, the biological activity of a new or untested chemical can be inferred from the molecular structure of similar compounds whose activities have already been assessed. QSARs attempt to relate physical and chemical properties of molecules to their biological activities. For this there are so many descriptors (for example, molecular weight, number of rotatable bonds, Log P) and simple statistical methods such as Multiple Linear Regression (MLR) are used to predict a model. These models describe the activity of the data set and can predict activities for further sets of (untested) compounds. These types of descriptors are simple to calculate and allow for a relatively fast analysis. 3D-QSAR uses probe-based sampling within a molecular lattice to determine three-dimensional properties of molecules (particularly steric and electrostatic values) and can then correlate these 3D descriptors with biological activity. Physicochemical descriptors, include hydrophobicity, topology, electronic properties, and steric effects etc. These descriptors can be calculated empirically, statistically or through more recent computational methods. QSARs are currently being applied in many disciplines, with many pertaining to drug design and environmental risk assessment.

show abstract

“…During the last ten years or so, a lot of progresses have been made in structure-based cheminformatics that are closely associated with the targets of medicinal chemistry, such as QSAR [1][2][3][4][5][6], molecular modeling and docking [7][8][9][10][11][12][13][14], as well as in biomedical cheminformatics such as identifying nucleosomes [15], predicting secretory proteins of malaria parasite [16], identifying HIV cleavage sites in proteins [17,18], identifying colorectal cancer related genes [19], predicting signal peptides [20], predicting protein subcellular locations [21,22], identifying DNA binding proteins [23], predicting proteases and their types [24], predicting antimicrobial peptides [25], identify nuclear receptors and their types [26], predicting GPCRs and their types [27], classifying hepatocellular cirrhosis and carcinoma [28], identifying anatomical therapeutic chemical (ATC) drugs classification [29], deciphering the effects of gene deletion on yeast longevity [30], and a series of powerful web-server predictors listed in Table 1, which is taken from the ref. 31 [31].…”

Section: Introductionmentioning

confidence: 99%

Investigation Binding Patterns of Human Carboxylesterase I (hCES I) with Broad Substrates by MD Simulations

Chu

Min²,

Zhang³

et al. 2013

CTMC

View full text Add to dashboard Cite

Human carboxylesterase I (hCES 1) plays an important role in the metabolism and activation of prodrugs, such as, the hydrolysis of a variety of drugs of prodrugs featuring an ester, amide or carbamate function. The bindings of the substrates of different lengths and cocaine to hCES1 at two different binding sites, catalytic site and Z-site, were studies through MD simulations. For each case, the correlation analysis has been performed to explore the binding patterns of a broad range of substrates binding to the hCES1.

show abstract

Multiple field three dimensional quantitative structure–activity relationship (MF‐3D‐QSAR)

Cited by 90 publications

References 33 publications

A Machine Learning Approach to Designing Guidelines for Acute Aquatic Toxicity

A Machine Learning Approach to Designing Guidelines for Acute Aquatic Toxicity

QSAR and its Role in Target-Ligand Interaction

Investigation Binding Patterns of Human Carboxylesterase I (hCES I) with Broad Substrates by MD Simulations

Contact Info

Product

Resources

About