Quantitative Structure Activity Relationship (QSAR) models can inform on the correlation between activities and structure-based molecular descriptors. This information is important for the understanding of the factors that govern molecular properties and for designing new compounds with favorable properties. Due to the large number of calculate-able descriptors and consequently, the much larger number of descriptors combinations, the derivation of QSAR models could be treated as an optimization problem. For continuous responses, metrics which are typically being optimized in this process are related to model performances on the training set, for example,

R^{2}

and

Q_{C V}^{2}

. Similar metrics, calculated on an external set of data (e.g.,

Q_{F 1 / F 2 / F 3}^{2}

), are used to evaluate the performances of the final models. A common theme of these metrics is that they are context -” ignorant”. In this work we propose that QSAR models should be evaluated based on their intended usage. More specifically, we argue that QSAR models developed for Virtual Screening (VS) should be derived and evaluated using a virtual screening-aware metric, e.g., an enrichment-based metric. To demonstrate this point, we have developed 21 Multiple Linear Regression (MLR) models for seven targets (three models per target), evaluated them first on validation sets and subsequently tested their performances on two additional test sets constructed to mimic small-scale virtual screening campaigns. As expected, we found no correlation between model performances evaluated by “classical” metrics, e.g.,

R^{2}

and

Q_{F 1 / F 2 / F 3}^{2}

and the number of active compounds picked by the models from within a pool of random compounds. In particular, in some cases models with favorable

R^{2}

and/or

Q_{F 1 / F 2 / F 3}^{2}

values were unable to pick a single active compound from within the pool whereas in other cases, models with poor

R^{2}

and/or

Q_{F 1 / F 2 / F 3}^{2}

values performed well in the context of virtual screening. We also found no significant correlation between the number of active compounds correctly identified by the models in the training, validation and test sets. Next, we have developed a new algorithm for the derivation of MLR models by optimizing an enrichment-based metric and tested its performances on the same datasets. We found that the best models derived in this manner showed, in most cases, much more consistent results across the training, validation and test sets and outperformed the corresponding MLR models in most virtual screening tests. Finally, we demonstrated that when tested as binary classifiers, models derived for the same targets by the new algorithm outperformed Random Forest (RF) and Support Vector Machine (SVM)-based models across training/validation/test sets, in most cases. We attribute the better performances of the Enrichment Optimizer Algorithm (EOA) models in VS to better handling of inactive random compounds. Optimizing an enrichment-based metric is therefore a promising strategy for the derivation of QSAR models for classification and virtual screening.

1,111 Results Found

Giving Molecules an Identity. On the Interplay Between QSARs and Partial Order Ranking

A DFT-Based QSARs Study of Acetazolamide/Sulfanilamide Derivatives with Carbonic Anhydrase (CA-II) Isozyme Inhibitory Activity

In Silico Prediction of Cytochrome P450-Drug Interaction: QSARs for CYP3A4 and CYP2C9

Recent Advances in Fragment-Based QSAR and Multi-Dimensional QSAR Methods

Combined Micellar Liquid Chromatography Technique and QSARs Modeling in Predicting the Blood–Brain Barrier Permeation of Heterocyclic Drug-like Compounds

QSPR/QSAR: State-of-Art, Weirdness, the Future

A QSAR Study for Antileishmanial 2-Phenyl-2,3-dihydrobenzofurans †

4D-QSAR: Perspectives in Drug Design

Chemical Graph Theory for Property Modeling in QSAR and QSPR—Charming QSAR & QSPR

Evaluation of QSAR Equations for Virtual Screening

QSAR Studies on Andrographolide Derivatives as α-Glucosidase Inhibitors

Enhanced QSAR Model Performance by Integrating Structural and Gene Expression Information

www.3d-qsar.com: A Portal to Build 3-D QSAR Models

Predictive QSAR Models for the Toxicity of Disinfection Byproducts

QSAR Models for the Prediction of Dietary Biomagnification Factor in Fish

QSAR Modeling on Benzo[c]phenanthridine Analogues as Topoisomerase I Inhibitors and Anti-cancer Agents

Multi-Strategy Assessment of Different Uses of QSAR under REACH Analysis of Alternatives to Advance Information Transparency

Two Decades of 4D-QSAR: A Dying Art or Staging a Comeback?

A QSAR Toxicity Study of a Series of Alkaloids with the Lycoctonine Skeleton

Study of the Applicability Domain of the QSAR Classification Models by Means of the Rivality and Modelability Indexes

Identification of Anti-SARS-CoV-2 Compounds from Food Using QSAR-Based Virtual Screening, Molecular Docking, and Molecular Dynamics Simulation Analysis

Combined Pharmacophore Modeling, Docking, and 3D-QSAR Studies of PLK1 Inhibitors

Toxicity Rank Order (TRO) As a New Approach for Toxicity Prediction by QSAR Models

Triazoloquinazolines as Human A3 Adenosine Receptor Antagonists: A QSAR Study

QSAR Study on 6-Substituted Benzimidazoles: An Insight into the Structural Requirement for the Angiotensin II AT1 Receptor Antagonist

(Q)SAR Models of HIV-1 Protein Inhibition by Drug-Like Compounds

QSAR Studies on N-aryl Derivative Activity Towards Alzheimer’s Disease

Understanding the Molecular Basis of 5-HT4 Receptor Partial Agonists through 3D-QSAR Studies

QSAR Analysis of 2-Amino or 2-Methyl-1-Substituted Benzimidazoles Against Pseudomonas aeruginosa

Inductive QSAR Descriptors. Distinguishing Compounds with Antibacterial Activity by Artificial Neural Networks

Synthesis, Cytotoxic Activity and 2D-QSAR Study of Some Imidazoquinazoline Derivatives

Self-organizing Neural Networks for Modeling Robust 3D and 4D QSAR: Application to Dihydrofolate Reductase Inhibitors

Quantitative Structure-Activity Relationship (QSAR) Studies on the Toxic Effects of Nitroaromatic Compounds (NACs): A Systematic Review

A (Comprehensive) Review of the Application of Quantitative Structure–Activity Relationship (QSAR) in the Prediction of New Compounds with Anti-Breast Cancer Activity

VLA-SMILES: Variable-Length-Array SMILES Descriptors in Neural Network-Based QSAR Modeling

BCL::EMAS — Enantioselective Molecular Asymmetry Descriptor for 3D-QSAR

Alert-QSAR. Implications for Electrophilic Theory of Chemical Carcinogenesis

Exploring the Prominent and Concealed Inhibitory Features for Cytoplasmic Isoforms of Hsp90 Using QSAR Analysis

Hybrid Classification/Regression Approach to QSAR Modeling of Stoichiometric Antiradical Capacity Assays’ Endpoints

QSAR Regression Models for Predicting HMG-CoA Reductase Inhibition

Pharmacophore Identification and QSAR Studies on Substituted Benzoxazinone as Antiplatelet Agents: kNN-MFA Approach

QSAR and Classification Study on Prediction of Acute Oral Toxicity of N-Nitroso Compounds

On Two Novel Parameters for Validation of Predictive QSAR Models

Design of New Derivatives of Dimedone Molecules Using QSAR and Docking Molecular

Rational Design of a Low-Data Regime of Pyrrole Antioxidants for Radical Scavenging Activities Using Quantum Chemical Descriptors and QSAR with the GA-MLR and ANN Concepts

New QSAR Models to Predict Human Transthyretin Disruption by Per- and Polyfluoroalkyl Substances (PFAS): Development and Application

Validation of Quantitative Structure-Activity Relationship (QSAR) Model for Photosensitizer Activity Prediction

A QSAR Study of Environmental Estrogens Based on a Novel Variable Selection Method

Mechanistic and Predictive QSAR Analysis of Diverse Molecules to Capture Salient and Hidden Pharmacophores for Anti-Thrombotic Activity

Novel 3-Amino-6-chloro-7-(azol-2 or 5-yl)-1,1-dioxo-1,4,2-benzodithiazine Derivatives with Anticancer Activity: Synthesis and QSAR Study

A QSAR Study for Antileishmanial 2-Phenyl-2,3-dihydrobenzofurans ^†

Understanding the Molecular Basis of 5-HT₄ Receptor Partial Agonists through 3D-QSAR Studies