Application of the “EigenValue Analysis (EVANS)” Methodology to Build Quantitative Structure Pharmacokinetic Relationship Models

We present EigenValue ANalySis (EVANS), a QSPR methodology that considers 3D molecular information of enantiomeric ensembles of chiral molecules without the need to perform an alignment step. EVANS follows an intricate molecular modelling protocol that generates orthogonal eigenvalues from hybrid matrices of physicochemical properties and 3D structure; these eigenvalues are used as independent variables in QSPR analyses. The EVANS formalism has been presented and deployed to build quantitative structure pharmacokinetic relationship (QSPKR) models on a benchmark dataset for three critical PK parameters: steady-state volume of distribution (VDss), clearance (CL), and half-life (t1/2). Predictive QSPKR models were built by using the eigenvalues generated via the EVANS methodology in conjunction with multiple linear regression (MLR), random forest (RF), and support vector machine (SVM) algorithms, and it was observed that the EVANS QSPKR models sync with published work in the literature. Thus, we present the EVANS methodology as a first-line prediction tool to prioritise compounds in drug discovery and development.

Download Full-text

Quantitative structure-pharmacokinetic relationship (QSPkP) analysis of the volume of distribution values of anti-infective agents from j group of the ATC classification in humans

Acta Pharmaceutica ◽

10.2478/v10007-012-0024-z ◽

2012 ◽

Vol 62 (3) ◽

pp. 305-323 ◽

Cited By ~ 3

Author(s):

Bruno Louis ◽

Vijay K. Agrawal

Keyword(s):

Statistical Tests ◽

External Validation ◽

Volume Of Distribution ◽

Support Vector ◽

Quantitative Structure ◽

Beta Lactam ◽

Svm Model ◽

Correlation Based Feature Selection ◽

Artificial Neural Network Ann ◽

D Values

In this study, a quantitative structure-pharmacokinetic relationship (QSPkR) model for the volume of distribution (Vd) values of 126 anti-infective drugs in humans was developed employing multiple linear regression (MLR), artificial neural network (ANN) and support vector regression (SVM) using theoretical molecular structural descriptors. A correlation-based feature selection (CFS) was employed to select the relevant descriptors for modeling. The model results show that the main factors governing Vd of anti-infective drugs are 3D molecular representations of atomic van der Waals volumes and Sanderson electronegativities, number of aliphatic and aromatic amino groups, number of beta-lactam rings and topological 2D shape of the molecule. Model predictivity was evaluated by external validation, using a variety of statistical tests and the SVM model demonstrated better performance compared to other models. The developed models can be used to predict the Vd values of anti-infective drugs.

Download Full-text

Development of a Hierarchical Support Vector Regression-Based In Silico Model for Caco-2 Permeability

Pharmaceutics ◽

10.3390/pharmaceutics13020174 ◽

2021 ◽

Vol 13 (2) ◽

pp. 174 ◽

Cited By ~ 1

Author(s):

Giang Huong Ta ◽

Cin-Syong Jhang ◽

Ching-Feng Weng ◽

Max K. Leong

Keyword(s):

Drug Discovery ◽

Support Vector Regression ◽

Human Colon ◽

Qsar Model ◽

Quantitative Structure Activity Relationship ◽

Support Vector ◽

Human Colon Carcinoma Cell ◽

Drug Discovery And Development ◽

Training Samples ◽

Experimental Values

Drug absorption is one of the critical factors that should be taken into account in the process of drug discovery and development. The human colon carcinoma cell layer (Caco-2) model has been frequently used as a surrogate to preliminarily investigate the intestinal absorption. In this study, a quantitative structure–activity relationship (QSAR) model was generated using the innovative machine learning-based hierarchical support vector regression (HSVR) scheme to depict the exceedingly confounding passive diffusion and transporter-mediated active transport. The HSVR model displayed good agreement with the experimental values of the training samples, test samples, and outlier samples. The predictivity of HSVR was further validated by a mock test and verified by various stringent statistical criteria. Consequently, this HSVR model can be employed to forecast the Caco-2 permeability to assist drug discovery and development.

Download Full-text

Support Vector Machine-Based Quantitative Structure-Activity Relationship Study of Cholesteryl Ester Transfer Protein Inhibitors

Chemical Biology & Drug Design ◽

10.1111/j.1747-0285.2009.00800.x ◽

2009 ◽

Vol 73 (5) ◽

pp. 558-571 ◽

Cited By ~ 12

Author(s):

Siavash Riahi ◽

Eslam Pourbasheer ◽

Mohammad Reza Ganjali ◽

Parviz Norouzi

Keyword(s):

Support Vector Machine ◽

Cholesteryl Ester ◽

Cholesteryl Ester Transfer Protein ◽

Quantitative Structure Activity Relationship ◽

Activity Relationship ◽

Support Vector ◽

Quantitative Structure ◽

Transfer Protein ◽

Structure Activity ◽

Relationship Study

Download Full-text

HUBUNGAN KUANTITATIF STRUKTUR TOKSISITAS : REVIEW

Pharmacoscript ◽

10.36423/pharmacoscript.v3i2.494 ◽

2020 ◽

Vol 3 (2) ◽

pp. 107-126

Author(s):

Purwaniati Purwaniati

Keyword(s):

In Silico ◽

Toxicity Data ◽

Quantitative Structure ◽

Training Set ◽

Drug Discovery And Development ◽

Independent Variables ◽

Drug Molecules ◽

Structure Toxicity Relationship ◽

Toxicity Relationship ◽

In Silico Methods

AbstrakProses penemuan dan pengembangan obat merupakan proses panjang yang memerlukan banyak waktu dan biaya. Ada banyak calon molekul obat yang gagal mencapai pasaran karena alasan toksisitasnya yang tinggi, sehingga harus dapat diidentifikasi sedini mungkin. Hubungan kuantitatif struktur toksisitas (HKST) merupakan salah satu metode in silico yang cukup tangguh untuk memprediksi toksisitas. HKST merupakan persamaan matematis yang dibentuk dari variabel data endpoint toksisitas seperti LD50 sebagai variabel terikat dan sejumlah deskriptor sebagai variable bebas yang dihitung dari senyawa-senyawa dalam training set. Persamaan HKST kemudian digunakan untuk memprediksi toksisitas senyawa baru.Kata kunci : toksisitas, hubungan kuantitatif struktur toksisitas (HKST)AbstractThe process of drug discovery and development is a long process that requires a lot of time and costly. There are many prospective drug molecules that fail to reach the market due to high toxicity reasons, so they must be identified as early as possible. The quantitative structure toxicity relationship (QSTR) is one of the in silico methods that is strong enough to predict toxicity. QSTR is a mathematical equation formed from endpoint toxicity data variables such as LD50 as a bound variable and a number of descriptors as independent variables calculated from the compounds in the training set. The QSTR equation is then used to predict the toxicity of new compounds.Keywords: toxicity, quantitative structure toxicity relationship (QSTR)

Download Full-text

Graph Mining in Chemoinformatics

Chemoinformatics and Advanced Machine Learning Perspectives ◽

10.4018/978-1-61520-911-8.ch006 ◽

2011 ◽

pp. 95-128 ◽

Cited By ~ 1

Author(s):

Hiroto Saigo ◽

Koji Tsuda

Keyword(s):

Graph Mining ◽

Chemical Compounds ◽

Quantitative Structure Activity Relationship ◽

Support Vector ◽

Quantitative Structure ◽

Least Squares Regression ◽

Lead Compounds ◽

Least Angle Regression ◽

Attributed Graphs ◽

Vector Machines

In standard QSAR (Quantitative Structure Activity Relationship) approaches, chemical compounds are represented as a set of physicochemical property descriptors, which are then used as numerical features for classification or regression. However, standard descriptors such as structural keys and fingerprints are not comprehensive enough in many cases. Since chemical compounds are naturally represented as attributed graphs, graph mining techniques allow us to create subgraph patterns (i.e., structural motifs) that can be used as additional descriptors. In this chapter, the authors present theoretically motivated QSAR algorithms that can automatically identify informative subgraph patterns. A graph mining subroutine is embedded in the mother algorithm and it is called repeatedly to collect patterns progressively. The authors present three variations that build on support vector machines (SVM), partial least squares regression (PLS) and least angle regression (LARS). In comparison to graph kernels, our methods are more interpretable, thereby allows chemists to identify salient subgraph features to improve the druglikeliness of lead compounds.

Download Full-text

Identification of Young Stellar Object candidates in the Gaia DR2 x AllWISE catalogue with machine learning methods

Monthly Notices of the Royal Astronomical Society ◽

10.1093/mnras/stz1301 ◽

2019 ◽

Vol 487 (2) ◽

pp. 2522-2537 ◽

Cited By ~ 12

Author(s):

G Marton ◽

P Ábrahám ◽

E Szegedi-Elek ◽

J Varga ◽

M Kun ◽

...

Keyword(s):

Machine Learning ◽

3D Structure ◽

Physical Nature ◽

Machine Learning Techniques ◽

Young Stellar Objects ◽

Support Vector ◽

Wide Field ◽

Stellar Objects ◽

Star Forming ◽

Gaia Dr2

ABSTRACT The second Gaia Data Release (DR2) contains astrometric and photometric data for more than 1.6 billion objects with mean Gaia G magnitude <20.7, including many Young Stellar Objects (YSOs) in different evolutionary stages. In order to explore the YSO population of the Milky Way, we combined the Gaia DR2 data base with Wide-field Infrared Survey Explorer (WISE) and Planck measurements and made an all-sky probabilistic catalogue of YSOs using machine learning techniques, such as Support Vector Machines, Random Forests, or Neural Networks. Our input catalogue contains 103 million objects from the DR2xAllWISE cross-match table. We classified each object into four main classes: YSOs, extragalactic objects, main-sequence stars, and evolved stars. At a 90 per cent probability threshold, we identified 1 129 295 YSO candidates. To demonstrate the quality and potential of our YSO catalogue, here we present two applications of it. (1) We explore the 3D structure of the Orion A star-forming complex and show that the spatial distribution of the YSOs classified by our procedure is in agreement with recent results from the literature. (2) We use our catalogue to classify published Gaia Science Alerts. As Gaia measures the sources at multiple epochs, it can efficiently discover transient events, including sudden brightness changes of YSOs caused by dynamic processes of their circumstellar disc. However, in many cases the physical nature of the published alert sources are not known. A cross-check with our new catalogue shows that about 30 per cent more of the published Gaia alerts can most likely be attributed to YSO activity. The catalogue can be also useful to identify YSOs among future Gaia alerts.

Download Full-text

Epothilones: Quantitative Structure Activity Relations Studied by Support Vector Machines and Artificial Neural Networks

QSAR & Combinatorial Science ◽

10.1002/qsar.200330837 ◽

2003 ◽

Vol 22 (7) ◽

pp. 722-728 ◽

Cited By ~ 15

Author(s):

Annalen Bleckmann ◽

Jens Meiler

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Support Vector Machines ◽

Support Vector ◽

Quantitative Structure ◽

Structure Activity ◽

Structure Activity Relations ◽

Vector Machines ◽

Artificial Neural

Download Full-text

TOPTMH: TOPOLOGY PREDICTOR FOR TRANSMEMBRANE α-HELICES

Journal of Bioinformatics and Computational Biology ◽

10.1142/s0219720010004501 ◽

2010 ◽

Vol 08 (01) ◽

pp. 39-57 ◽

Cited By ~ 4

Author(s):

REZWAN AHMED ◽

HUZEFA RANGWALA ◽

GEORGE KARYPIS

Keyword(s):

Markov Models ◽

Structural Characteristics ◽

Transmembrane Helix ◽

3D Structure ◽

Prediction Method ◽

Support Vector ◽

Svm Classifier ◽

Sequence Information ◽

Vector Machines ◽

Prediction Approach

Alpha-helical transmembrane proteins mediate many key biological processes and represent 20%–30% of all genes in many organisms. Due to the difficulties in experimentally determining their high-resolution 3D structure, computational methods to predict the location and orientation of transmembrane helix segments using sequence information are essential. We present TOPTMH, a new transmembrane helix topology prediction method that combines support vector machines, hidden Markov models, and a widely used rule-based scheme. The contribution of this work is the development of a prediction approach that first uses a binary SVM classifier to predict the helix residues and then it employs a pair of HMM models that incorporate the SVM predictions and hydropathy-based features to identify the entire transmembrane helix segments by capturing the structural characteristics of these proteins. TOPTMH outperforms state-of-the-art prediction methods and achieves the best performance on an independent static benchmark.

Download Full-text