Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Mar 16, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

Large-scale ligand-based predictive modelling using support vector machines.

Jonathan Alvarsson¹, Samuel Lampa¹, Wesley Schaal²

¹Department of Pharmaceutical Biosciences, Uppsala University, 751 24 Uppsala, Sweden.

Journal of Cheminformatics

|August 13, 2016

Summary

This summary is machine-generated.

Related Concept Videos

Ligand Binding Sites

Ligand Binding Sites

Proteins are dynamic macromolecules that carry out a wide variety of essential processes; however, the activities of most proteins depend on their interactions with other molecules or ions, known as ligands.
Protein-ligand interactions are quite specific; even though numerous potential ligands surround a cellular protein at any given time, only a particular ligand can bind to that protein. Moreover, a ligand binds only to a dedicated area on the surface of the protein, known as the...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

High Prevalences of Current Mental Disorder Diagnoses Among University Students in Sweden: A Cross-Sectional Study With Implications for Student Mental Health Services.

International journal of methods in psychiatric research·2026

Same author

Cell painting and thermal proteome profiling for inference of drug targets and mechanism of action.

Molecular systems biology·2026

Same author

Demic diffusion can create chimpanzee gene-cultural linkage.

Journal of human evolution·2026

Same author

An integrative molecular map of pediatric B-cell precursor acute lymphoblastic leukemia.

Communications medicine·2026

Same author

App-based skills training to reduce problem drinking among adult internet help-seekers: a double-blinded randomized controlled trial.

Addiction science & clinical practice·2026

Same author

University students' interest in intervention research participation is associated with gender, age and symptoms of depression, general and social anxiety, panic disorder and self-harm.

BMC research notes·2026

Same journal

Unified heterogeneity-aware benchmark of drug synergy prediction: a cross-study analysis of traditional machine learning and graph deep learning models.

Journal of cheminformatics·2026

Same journal

Count your bits: fingerprint benchmarking to assess broad chemical space representation.

Journal of cheminformatics·2026

Same journal

Sampling out-of-distribution chemical spaces via Bayesian flow.

Journal of cheminformatics·2026

Same journal

Hold on tight: the kinetic profiling of opioid receptor ligands using the CORAL-MD.

Journal of cheminformatics·2026

Same journal

Transformer-accelerated discovery of inhibitors targeting the RpsA<sub>Δ438</sub> deletion in PZA-resistant tuberculosis.

Journal of cheminformatics·2026

Same journal

DICL: a manually curated database of ion channels and ligands as a useful platform for drug discovery targeting ion channels.

Journal of cheminformatics·2026

See all related articles

LIBLINEAR support vector machines (SVM) offer efficient predictive modeling for large drug discovery datasets, outperforming traditional methods in speed and comparable accuracy. Non-linear models were infeasible for large-scale data.

Area of Science:

Computational chemistry
Cheminformatics
Drug discovery modeling

Background:

Increasing dataset sizes in drug discovery pose challenges for predictive model development time and accuracy.
Ligand-based regression models are crucial for predicting molecular properties.
Efficient computational methods are needed to handle large chemical structure datasets.

Purpose of the Study:

To investigate the impact of dataset size on predictive model performance and computational time.
To compare different support vector machine (SVM) implementations for large-scale chemical data modeling.
To evaluate the feasibility of non-linear kernels with large datasets.

Main Methods:

Trained ligand-based regression models on open datasets up to 1.2 million chemical structures.

Keywords:

Bioclipse Molecular signatures Predictive modelling QSAR Support vector machine

Related Experiment Videos

Last Updated: Mar 16, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

Utilized two support vector machine (SVM) implementations: LIBLINEAR and libsvm with a radial basis function kernel.

Employed the signatures molecular descriptor to represent chemical structures.

Main Results:

LIBLINEAR SVM demonstrated comparable predictive performance to libsvm on larger datasets.
LIBLINEAR significantly reduced model building time, even on modest computational resources.
Non-linear kernel SVMs were computationally infeasible for large dataset sizes, even with cluster computing.

Conclusions:

LIBLINEAR SVM is an efficient and accurate method for building predictive models from large drug discovery datasets.
Linear SVMs are more suitable than non-linear SVMs for very large chemical datasets.
Extended Bioclipse framework to support LIBLINEAR models for logD and solubility predictions.