Related Experiment Video
Updated: Jan 23, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
Fitness of unregulated human Ras mutants modeled by implementing computational mutagenesis and machine learning
Majid Masso1, Arnav Bansal1, Preethi Prem1
1School of Systems Biology, George Mason University, 10900 University Blvd. MS 5B3, Manassas, Virginia, 20110, USA.
Researchers developed AI models to predict H-Ras protein fitness for mutants, aiding oncogene research and potential cancer treatment design.
Area of Science:
- Molecular Biology
- Computational Biology
- Oncogenesis
Background:
- Ras proteins are key oncogenes regulating cell growth, differentiation, and proliferation.
- Understanding mutations in Ras proteins is crucial for cancer research and therapeutic development.
Purpose of the Study:
- To develop predictive models for H-Ras protein fitness using experimental data and AI.
- To assess the accuracy of computational mutagenesis and machine learning algorithms in predicting protein variant fitness.
- To explore the potential of these models in understanding oncogenesis and designing cancer therapies.
Main Methods:
- Utilized experimental fitness data and computational mutagenesis to generate feature vectors of protein structural changes.
- Implemented machine learning algorithms, including random forest classification and tree regression.
- Employed cross-validation and control experiments to evaluate model performance and statistical significance.
Main Results:
- Developed models that reliably predict fitness for single residue mutants of H-Ras p21.
- Achieved high accuracy in classification models (balanced accuracy up to 82%, MCC 0.63, AUC 0.90) and regression models (Pearson's correlation 0.79).
- Demonstrated superior performance compared to state-of-the-art methods and random guessing on control data.
Conclusions:
- The study presents a successful proof-of-principle for using AI and computational mutagenesis to predict protein fitness.
- The developed models offer a complementary approach to understanding oncogenesis mechanisms in Ras proteins and related isoforms.
- Findings provide valuable insights for designing future diagnostic and therapeutic strategies for cancers involving Ras mutations.
More Related Videos
07:15Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
Related Concept Videos
Induced-fit Model
Enzymes exhibit substrate specificity, meaning that they can only bind to certain substrates. This is mainly determined by the shape and chemical...
The Ras Gene
Ras is a...
In-vitro Mutagenesis
Small GTPases - Ras and Rho
Three regulatory proteins control their activity:
Simplified Synchronous Machine Model
In this model, each generator is connected to a...
Wind Turbine Machine Models
Induction machines interact through the rotating magnetic field generated by the stator and the rotor. The key parameter is slip, which is the difference between synchronous speed and rotor speed relative to synchronous speed. Slip is...