Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Ribosome Profiling02:24

Ribosome Profiling

Ribosome profiling or ribo-sequencing is a deep sequencing technique that produces a snapshot of active translation in a cell. It selectively sequences the mRNAs protected by ribosomes to get an insight into a cell’s translation landscape at any given point in time.
Applications of ribosome profiling
Ribosome profiling has many applications, including in vivo monitoring of translation inside a particular organ or tissue type and quantifying new protein synthesis levels.
The technique helps...

You might also read

Related Articles

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

Sort by
Same author

Catalytic Asymmetric Hydration of Alkenes.

Journal of the American Chemical Society·2026
Same author

Predicting Enantioselectivity via Kinetic Simulations on Gigantic Reaction Path Networks.

ACS central science·2026
Same author

Current Insights on Skin Permeability Data and Quantitative Structure-Property Relationship Modeling.

Molecular informatics·2026
Same author

Interpretable and Scalable Similarity Metrics for DNA-Encoded Library Design Using Generative Topographic Mapping.

Molecular informatics·2026
Same author

Toward Reaction Vessel Mimicry: Machine Learning-Assisted Automated Exploration of Alkene Polymerization and Its Transferability.

Journal of chemical theory and computation·2026
Same author

An Accurate and Efficient Reaction Path Search with Iteratively Trained Neural Network Potential: Answering the Passerini Mechanism Controversy.

Journal of chemical theory and computation·2025

Related Experiment Video

Updated: Jul 8, 2026

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins
05:08

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Published on: July 8, 2025

In Silico ADMET: From Current Practices to Novel Profilers.

Pierre Llompart1,2, Claire Minoletti2, Gilles Marcou1

  • 1Laboratory of Chemoinformatics, UMR7140, University of Strasbourg, Strasbourg 67000, France.

Journal of Medicinal Chemistry
|July 7, 2026
PubMed
Summary

Multitask learning (MTL) in drug discovery improves predictions using curated data. A new unified model and dataset (OneADMET) show MTL matches or exceeds single-task performance for ADMET profiling.

More Related Videos

Probing RNA Structure with Dimethyl Sulfate Mutational Profiling with Sequencing In Vitro and in Cells
10:34

Probing RNA Structure with Dimethyl Sulfate Mutational Profiling with Sequencing In Vitro and in Cells

Published on: December 9, 2022

Related Experiment Videos

Last Updated: Jul 8, 2026

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins
05:08

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Published on: July 8, 2025

Probing RNA Structure with Dimethyl Sulfate Mutational Profiling with Sequencing In Vitro and in Cells
10:34

Probing RNA Structure with Dimethyl Sulfate Mutational Profiling with Sequencing In Vitro and in Cells

Published on: December 9, 2022

Area of Science:

  • Computational drug discovery
  • Pharmacokinetics and ADMET profiling
  • Machine learning in cheminformatics

Background:

  • Multitask learning (MTL) offers improved predictive performance and generalization in computational drug discovery compared to single-task models.
  • Existing ADMET (absorption, distribution, metabolism, elimination, and toxicity) web services often use uncurated, redundant data, limiting their reliability.
  • There is a need for curated datasets and advanced models to enhance drug design predictions.

Purpose of the Study:

  • To critically review existing open-source ADMET web services and identify limitations in data curation and diversity.
  • To introduce OneADMET, a large-scale, meticulously curated dataset for ADMET and biological activity prediction.
  • To develop and evaluate a unified multitask learning model for efficient and accurate prediction of numerous ADMET endpoints.

Main Methods:

  • Critical review of open-source ADMET web services and their underlying datasets.
  • Creation of the OneADMET dataset, comprising 1,119,719 measurements for 738,161 compounds across 44 ADMET endpoints and 1,489 biological activities.
  • Development of a unified ChemProp-based multitask learning (MTL) model capable of handling hundreds of continuous tasks simultaneously.

Main Results:

  • The review revealed significant data redundancy and limited curation in existing ADMET web services.
  • The OneADMET dataset provides a comprehensive and high-quality resource for computational drug discovery.
  • The unified MTL model demonstrated comparable or superior predictive accuracy to single-task models across various ADMET endpoints.
  • MTL models offer practical advantages in deployment and maintenance for large-scale prediction tasks.

Conclusions:

  • Large-scale multitask learning is highly effective for pharmacokinetics and ADMET profiling in drug discovery.
  • The OneADMET dataset and the unified MTL model represent valuable contributions to the computational drug discovery community.
  • This work underscores the importance of data curation and advanced modeling techniques for reliable drug design predictions.