Related Experiment Video
Updated: Jul 18, 2025
Network Pharmacology Prediction and Metabolomics Validation of the Mechanism of Fructus Phyllanthi against Hyperlipidemia
Published on: April 7, 2023
Discovering the mechanism of action of drugs with a sparse explainable network
Katyna Sada Del Real1, Angel Rubio2
1Departamento de Ingeniería Biomédica y Ciencias, TECNUN, Universidad de Navarra, San Sebastián 20018, Spain.
Background:
Although Deep Neural Networks (DDNs) have been successful in predicting the efficacy of cancer drugs, the lack of explainability in their decision-making process is a significant challenge. Previous research proposed mimicking the Gene Ontology structure to allow for interpretation of each neuron in the network. However, these previous approaches require huge amount of GPU resources and hinder its extension to genome-wide models.
Methods:
We developed SparseGO, a sparse and interpretable neural network, for predicting drug response in cancer cell lines and their Mechanism of Action (MoA). To ensure model generalization, we trained it on multiple datasets and evaluated its performance using three cross-validation schemes. Its efficiency allows it to be used with gene expression. In addition, SparseGO integrates an eXplainable Artificial Intelligence (XAI) technique, DeepLIFT, with Support Vector Machines to computationally discover the MoA of drugs.
Findings:
SparseGO's sparse implementation significantly reduced GPU memory usage and training speed compared to other methods, allowing it to process gene expression instead of mutations as input data. SparseGO using expression improved the accuracy and enabled its use on drug repositioning. Furthermore, gene expression allows the prediction of MoA using 265 drugs to train it. It was validated on understudied drugs such as parbendazole and PD153035.
Interpretation:
SparseGO is an effective XAI method for predicting, but more importantly, understanding drug response.
Funding:
The Accelerator Award Programme funded by Cancer Research UK [C355/A26819], Fundación Científica de la AECC and Fondazione AIRC, Project PIBA_2020_1_0055 funded by the Basque Government and the Synlethal Project (RETOS Investigacion, Spanish Government).
Insights
SparseGO, a novel sparse and interpretable neural network, enhances cancer drug response prediction and understanding. This explainable AI method improves accuracy and efficiency, enabling drug repositioning and Mechanism of Action discovery.
Area of Science:
- Computational biology
- Artificial intelligence in oncology
- Drug discovery and development
Background:
- Deep neural networks (DNNs) show promise in predicting cancer drug efficacy but lack explainability.
- Previous interpretation methods require substantial GPU resources and limit genome-wide applications.
- Explainable AI (XAI) is crucial for understanding complex biological models.
Purpose of the Study:
- To develop a sparse and interpretable neural network for predicting cancer drug response and Mechanism of Action (MoA).
- To improve the efficiency and scalability of DNNs for genomic data analysis in drug discovery.
- To integrate XAI techniques for discovering novel drug MoAs.
Main Methods:
- Developed SparseGO, a sparse and interpretable neural network.
- Integrated DeepLIFT (XAI) with Support Vector Machines for MoA discovery.
- Trained and evaluated SparseGO on multiple datasets using cross-validation, utilizing gene expression data.
Main Results:
- SparseGO significantly reduced GPU memory usage and training time compared to existing methods.
- Using gene expression as input improved prediction accuracy and enabled drug repositioning.
- Successfully predicted MoA for 265 drugs and validated on understudied compounds.
Conclusions:
- SparseGO is an effective XAI method for predicting and understanding cancer drug response.
- The model's efficiency and interpretability facilitate broader applications in precision oncology.
- SparseGO advances the field of explainable AI in biomedical research.
More Related Videos
13:18Network Pharmacology Prediction and Experimental Validation of Trichosanthes-Fritillaria thunbergii Action Mechanism Against Lung Adenocarcinoma
Published on: March 3, 2023
12:34Author Spotlight: Exploring ShiDuGao's Multi-Target Approach in Anus Eczema Treatment
Published on: January 12, 2024
Related Concept Videos
01:24Principles of Drug Action
Drugs can be agonists or antagonists. Like the endogenous ligands, agonists always bind and activate the target to produce a cellular response. Agonist binding induces a conformational change which in turn...
01:26Targets for Drug Action: Overview
Receptors are either membrane-spanning or intracellular proteins, which upon binding a ligand, get activated and transmit the signal downstream to elicit a response. Drugs bind receptors, either mimicking the action of endogenous ligands or blocking the receptor activity to bring about a modified response. Nearly 35% of approved drugs target the G...
01:21Pharmacodynamics: Overview and Principles
Most drugs' effects result from their interactions with drug receptors or targets within the body. These interactions trigger specific responses at the cellular or systemic level. Drug receptors can be found on the surfaces of cells or...
01:26Neurochemical Transmission: Sites of Drug Action
01:29Drug-Receptor Interactions
Several parameters, such as the drug's affinity for its receptor and its efficacy, which is its ability to activate the receptor, determine the drug's effect on the tissue....
01:21Mechanistic Models: Overview of Compartment Models