Ligand Classifier of Adaptively Boosting Ensemble Decision Stumps (LiCABEDS) and Its Application on Modeling Ligand Functionality for 5HT-Subtype GPCR Families

Abstract

Advanced high-throughput screening (HTS) technologies generate great amounts of bioactivity data, and this data needs to be analyzed and interpreted with attention to understand how these small molecules affect biological systems. As such, there is an increasing demand to develop and adapt cheminformatics algorithms and tools in order to predict molecular and pharmacological properties based on these large datasets. In this manuscript, we report a novel machine-learningbased ligand classification algorithm, named Ligand Classifier of Adaptively Boosting Ensemble Decision Stumps (LiCABEDS), for data-mining and modeling of large chemical datasets to predict pharmacological properties in an efficient and accurate manner. The performance of LiCABEDS was evaluated through predicting GPCR ligand functionality (agonist or antagonist) using four different molecular fingerprints, including Maccs, FP2, Unity and Molprint 2D fingerprints. Our studies showed that LiCABEDS outperformed two other popular techniquesclassification tree and Naive Bayes classifier -on all four types of molecular fingerprints. Parameters in LiCABEDS, including the number of boosting iterations, initialization condition, and a "reject option" boundary, were thoroughly explored and discussed to demonstrate the capability of handling imbalanced datasets, as well as its robustness and flexibility. In addition, the detailed mathematical concepts and theory are also given to address the principle behind statistical prediction models. The LiCABEDS algorithm has been implemented into a user-friendly software package that is accessible online at http://www.cbligand.org/LiCABEDS/.

LiCABEDS II. Modeling of Ligand Selectivity for G-Protein-Coupled Cannabinoid Receptors

“…Parameter M was thoroughly discussed in our previous study 16 , and the hypothesis was that large M produced close-to-optimal models. Using cross-validation might improve prediction performance, but its effect was statistically insignificant.…”

“…The performance, robustness, interpretability and parameters of LiCABEDS were thoroughly discussed through modeling 5-HT 1A ligand functionality 16 . Now we aim to demonstrate the wide applicability of LiCABEDS by modeling cannabinoid ligand selectivity.…”

“…The whole cannabinoid ligand dataset produced 2530 three-layer Molprint 2D features from Bender et al's program. These features were mapped to binary vector according to previously published protocol 16 . To systematically evaluate prediction accuracy, 50% CB1 selective and 50% CB1 non-selective compounds were randomly selected as training set in order to build a prediction model.…”

“…Li gand C lassifier of A daptively B oosting E nsemble D ecision S tumps (LiCABEDS) is a generic ligand classification algorithm developed by Xie's laboratory 16 for the prediction of categorical ligand properties. LiCABEDS was established on Freund and Schapire's ensemble learning framework 17 .…”

Advances in Recent Patent and Clinical Trial Drug Development for Alzheimer’s Disease

“…Of course, it is a challenge to apply such multidrug remedies for AD treatment with clinical rationale. Thus, new approaches such as quantitative system pharmacology with computational system polypharmacology algorithms [157,158] and chemogenomics knowledgebases [159,160] will open up a broad and promising avenue to advance the discovery and development of new-generation drugs for AD in the future.…”