DualGCN: a dual graph convolutional network model to predict cancer drug response
Tianxing Ma1, Qiao Liu2, Haochen Li3
1MOE Key Laboratory of Bioinformatics, Bioinformatics Division, BNRIST and Department of Automation, Tsinghua University, Beijing, 100084, China.
Background:
Drug resistance is a critical obstacle in cancer therapy. Discovering cancer drug response is important to improve anti-cancer drug treatment and guide anti-cancer drug design. Abundant genomic and drug response resources of cancer cell lines provide unprecedented opportunities for such study. However, cancer cell lines cannot fully reflect heterogeneous tumor microenvironments. Transferring knowledge studied from in vitro cell lines to single-cell and clinical data will be a promising direction to better understand drug resistance. Most current studies include single nucleotide variants (SNV) as features and focus on improving predictive ability of cancer drug response on cell lines. However, obtaining accurate SNVs from clinical tumor samples and single-cell data is not reliable. This makes it difficult to generalize such SNV-based models to clinical tumor data or single-cell level studies in the future.
Results:
We present a new method, DualGCN, a unified Dual Graph Convolutional Network model to predict cancer drug response. DualGCN encodes both chemical structures of drugs and omics data of biological samples using graph convolutional networks. Then the two embeddings are fed into a multilayer perceptron to predict drug response. DualGCN incorporates prior knowledge on cancer-related genes and protein-protein interactions, and outperforms most state-of-the-art methods while avoiding using large-scale SNV data.
Conclusions:
The proposed method outperforms most state-of-the-art methods in predicting cancer drug response without the use of large-scale SNV data. These favorable results indicate its potential to be extended to clinical and single-cell tumor samples and advancements in precision medicine.
Insights
A new Dual Graph Convolutional Network (DualGCN) model predicts cancer drug response by analyzing drug chemical structures and biological omics data. This method improves upon existing approaches by not relying on single nucleotide variants (SNV) data.
Area of Science:
- Oncology
- Bioinformatics
- Computational Biology
Background:
- Drug resistance poses a significant challenge in cancer therapy, necessitating improved methods for predicting anti-cancer drug responses.
- Cancer cell line data offers valuable insights but doesn't fully represent tumor microenvironments, highlighting the need to bridge in vitro findings with clinical and single-cell data.
- Current predictive models often rely on single nucleotide variants (SNVs), which are difficult to obtain reliably from clinical and single-cell samples, limiting their applicability.
Purpose of the Study:
- To develop a novel computational method for predicting cancer drug response that overcomes limitations of existing SNV-based approaches.
- To create a unified model that integrates drug chemical structures and biological sample omics data for enhanced predictive accuracy.
- To enable the generalization of drug response predictions from cell line studies to clinical and single-cell tumor data.
Main Methods:
- Introduced DualGCN, a Dual Graph Convolutional Network model for predicting cancer drug response.
- Encoded drug chemical structures and biological sample omics data using graph convolutional networks.
- Integrated prior knowledge of cancer-related genes and protein-protein interactions into the model.
Main Results:
- DualGCN effectively predicts cancer drug response by encoding both drug chemical structures and omics data.
- The model demonstrated superior performance compared to most state-of-the-art methods.
- DualGCN successfully avoided the need for large-scale single nucleotide variant (SNV) data, enhancing its applicability to diverse datasets.
Conclusions:
- The developed DualGCN method shows significant promise for predicting cancer drug response, outperforming existing approaches.
- The model's ability to function without SNV data makes it suitable for extension to clinical and single-cell tumor samples.
- These findings represent a potential advancement in precision medicine by improving anti-cancer drug treatment strategies.
More Related Videos
13:34A Combined 3D Tissue Engineered In Vitro/In Silico Lung Tumor Model for Predicting Drug Effectiveness in Specific Mutational Backgrounds
Published on: April 6, 2016
06:46Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
Related Concept Videos
02:56Treatment Resistant Cancers
01:21Cancer Survival Analysis
02:50Combination Therapies and Personalized Medicine
The combination of the drug acetazolamide and sulforaphane is a good example of combination therapy to treat cancer. The cells in the interior of a large tumor often die due to the hypoxic and...
02:43Mouse Models of Cancer Study
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
01:07Determination of Renal Drug Clearance: Graphical and Midpoint Methods
The graphical method involves plotting the rate of drug excretion in urine against the plasma drug concentration. By analyzing the graph, the clearance can be calculated and obtained. Drugs rapidly excreted by the kidneys exhibit a...