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

In vitro Mutagenesis01:16

In vitro Mutagenesis

To learn more about the function of a gene, researchers can observe what happens when the gene is inactivated or “knocked out,” by creating genetically engineered knockout animals. Knockout mice have been particularly useful as models for human diseases such as cancer, Parkinson’s disease, and diabetes.
Mutagenicity and Carcinogenicity01:25

Mutagenicity and Carcinogenicity

Mutagenicity and carcinogenicity refer to the ability of drugs to cause genetic defects and induce cancer, respectively. The International Agency for Research on Cancer (IARC) classifies agents into four groups based on their carcinogenic potential. Group 1 agents are known human carcinogens; group 2A agents are probably carcinogenic to humans; group 3 agents lack data to support their role in carcinogenesis; and group 4 includes agents for which data support that they are not likely to be...
In-vitro Mutagenesis01:16

In-vitro Mutagenesis

To learn more about the function of a gene, researchers can observe what happens when the gene is inactivated or “knocked out,” by creating genetically engineered knockout animals. Knockout mice have been particularly useful as models for human diseases such as cancer, Parkinson’s disease, and diabetes.
Spontaneous and Induced Mutations01:30

Spontaneous and Induced Mutations

Spontaneous mutations arise infrequently during DNA replication due to errors in the process. A key factor behind these errors is tautomeric shifts in nitrogenous bases, where bases transition from keto to enol forms or amino to imino forms. This shift can alter base-pairing rules, leading to mutations. Additionally, reactive oxygen species (ROS) arising from aerobic metabolism can damage DNA, resulting in depurination (loss of a purine base) or depyrimidination (loss of a pyrimidine base).

You might also read

Related Articles

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

Sort by
Same author

Structures of human protein tyrosine phosphatase variants reveal targetable allosteric sites.

The Journal of biological chemistry·2025
Same author

Orchestrating function: Concerted dynamics, allostery, and catalysis in protein tyrosine phosphatases.

Current opinion in structural biology·2025
Same author

Allosteric regulation of the tyrosine phosphatase PTP1B by a protein-protein interaction.

Protein science : a publication of the Protein Society·2024
Same author

Structures of human PTP1B variants reveal allosteric sites to target for weight loss therapy.

bioRxiv : the preprint server for biology·2024
Same author

Allosteric regulation of the tyrosine phosphatase PTP1B by a protein-protein interaction.

bioRxiv : the preprint server for biology·2024
Same author

Native dynamics and allosteric responses in PTP1B probed by high-resolution HDX-MS.

Protein science : a publication of the Protein Society·2024

Related Experiment Video

Updated: Jun 30, 2026

A High-throughput Assay for the Prediction of Chemical Toxicity by Automated Phenotypic Profiling of Caenorhabditis elegans
09:01

A High-throughput Assay for the Prediction of Chemical Toxicity by Automated Phenotypic Profiling of Caenorhabditis elegans

Published on: March 14, 2019

AmesNet: A Task-Conditioned Deep Learning Model with Enhanced Sensitivity and Generalization in Ames Mutagenicity

Tyler J Umansky1, Virgil A Woods1, Sean M Russell1

  • 1Model Medicines, San Diego, California 92122, United States.

Chemical Research in Toxicology
|June 29, 2026
PubMed
Summary

AmesNet improves in silico mutagenicity prediction for novel drug candidates, enhancing sensitivity and balanced accuracy on out-of-domain data. This AI model helps prevent costly late-stage failures by enabling earlier mutagenicity triage.

More Related Videos

The Lambda Select cII Mutation Detection System
07:08

The Lambda Select cII Mutation Detection System

Published on: April 26, 2018

Related Experiment Videos

Last Updated: Jun 30, 2026

A High-throughput Assay for the Prediction of Chemical Toxicity by Automated Phenotypic Profiling of Caenorhabditis elegans
09:01

A High-throughput Assay for the Prediction of Chemical Toxicity by Automated Phenotypic Profiling of Caenorhabditis elegans

Published on: March 14, 2019

The Lambda Select cII Mutation Detection System
07:08

The Lambda Select cII Mutation Detection System

Published on: April 26, 2018

Area of Science:

  • Computational toxicology
  • Drug discovery and development
  • Artificial intelligence in chemistry

Background:

  • Genotoxicity assessments are crucial for novel therapeutics but are often delayed due to cost and complexity.
  • Late-stage failures in drug development, often stemming from safety issues like mutagenicity, can result in significant financial and temporal losses.
  • Current in silico Ames mutagenicity models struggle with sensitivity for novel chemical structures, leading to potential false negatives and continued development of unsafe compounds.

Purpose of the Study:

  • To develop an advanced in silico Ames mutagenicity model, AmesNet, capable of high performance on novel chemical spaces.
  • To address the sensitivity drop-off issue in existing Ames models when evaluating out-of-domain compounds.
  • To provide a reliable early-stage mutagenicity triage tool for drug developers, mitigating late-stage failure risks.

Main Methods:

  • Developed AmesNet, a task-conditioned model with a dual-branch architecture incorporating a molecular encoder and an Ames assay context channel.
  • Incorporated specific Ames assay conditions, such as metabolic activation and bacterial strain type, into the model's conditioning.
  • Evaluated AmesNet's performance on an out-of-domain (OOD) test set from the Second Ames/QSAR International Challenge Project.

Main Results:

  • AmesNet achieved a sensitivity of 0.72 (95% CI: 0.68-0.76) and a balanced accuracy (BA) of 0.81 (95% CI: 0.78-0.83) on the OOD test data.
  • Outperformed existing models by improving discrimination without sacrificing overall performance, unlike methods that trade BA for sensitivity.
  • Structural analysis indicated AmesNet successfully identified mutagenic compounds missed by other models.

Conclusions:

  • AmesNet demonstrates superior performance in predicting Ames mutagenicity for novel chemical entities, particularly in out-of-domain scenarios.
  • The model's ability to maintain high sensitivity and balanced accuracy offers a significant improvement over current in silico approaches.
  • AmesNet provides a valuable high-confidence filtering mechanism, enabling earlier and more efficient mutagenicity triage in drug development pipelines.