Related Experiment Video
Updated: Jan 10, 2026
A Data Integration Workflow to Identify Drug Combinations Targeting Synthetic Lethal Interactions
Published on: May 27, 2021
Humanized yeast genetic interaction mapping predicts synthetic lethal interactions of FBXW7 in breast cancer
Morgan W B Kirzinger1, Frederick S Vizeacoumar2, Bjorn Haave2
1Department of Computer Science, College of Arts and Science, University of Saskatchewan, 176 Thorvaldson Bldg, 110 Science Place, Saskatoon, Saskatchewan, S7N 5C9, Canada.
Background:
Synthetic lethal interactions (SLIs) that occur between gene pairs are exploited for cancer therapeutics. Studies in the model eukaryote yeast have identified ~ 550,000 negative genetic interactions that have been extensively studied, leading to characterization of novel pathways and gene functions. This resource can be used to predict SLIs that can be relevant to cancer therapeutics.
Methods:
We used patient data to identify genes that are down-regulated in breast cancer. InParanoid orthology mapping was performed to identify yeast orthologs of the down-regulated genes and predict their corresponding SLIs in humans. The predicted network graphs were drawn with Cytoscape. CancerRXgene database was used to predict drug response.
Results:
Harnessing the vast available knowledge of yeast genetics, we generated a Humanized Yeast Genetic Interaction Network (HYGIN) for 1009 human genes with 10,419 interactions. Through the addition of patient-data from The Cancer Genome Atlas (TCGA), we generated a breast cancer specific subnetwork. Specifically, by comparing 1009 genes in HYGIN to genes that were down-regulated in breast cancer, we identified 15 breast cancer genes with 130 potential SLIs. Interestingly, 32 of the 130 predicted SLIs occurred with FBXW7, a well-known tumor suppressor that functions as a substrate-recognition protein within a SKP/CUL1/F-Box ubiquitin ligase complex for proteasome degradation. Efforts to validate these SLIs using chemical genetic data predicted that patients with loss of FBXW7 may respond to treatment with drugs like Selumitinib or Cabozantinib.
Conclusions:
This study provides a patient-data driven interpretation of yeast SLI data. HYGIN represents a novel strategy to uncover therapeutically relevant cancer drug targets and the yeast SLI data offers a major opportunity to mine these interactions.
Insights
Yeast genetic interactions predict potential cancer drug targets. This study created a humanized network to identify synthetic lethal interactions (SLIs) relevant to breast cancer, revealing potential drug responses for patients with specific gene losses.
Area of Science:
- Genetics
- Cancer Biology
- Bioinformatics
Background:
- Synthetic lethal interactions (SLIs) are crucial for cancer therapeutics.
- Yeast genetics provides a rich resource for identifying gene functions and pathways.
- Over 550,000 negative genetic interactions have been cataloged in yeast.
Purpose of the Study:
- To leverage yeast genetic interaction data for predicting human SLIs relevant to cancer.
- To develop a humanized yeast genetic interaction network (HYGIN) for cancer research.
- To identify potential therapeutic targets and drug responses in breast cancer.
Main Methods:
- Orthology mapping of human genes down-regulated in breast cancer to yeast genes.
- Construction of a humanized yeast genetic interaction network (HYGIN).
- Integration of patient data from The Cancer Genome Atlas (TCGA) to create a breast cancer-specific subnetwork.
- Utilizing the CancerRXgene database for drug response prediction.
Main Results:
- A humanized yeast genetic interaction network (HYGIN) was generated with 1009 human genes and 10,419 interactions.
- A breast cancer-specific subnetwork identified 15 genes with 130 potential SLIs.
- 32 SLIs involved FBXW7, a known tumor suppressor.
- Predicted that loss of FBXW7 may sensitize patients to Selumitinib or Cabozantinib.
Conclusions:
- This study presents a patient-data driven approach to interpret yeast SLI data.
- HYGIN offers a novel strategy for discovering cancer drug targets.
- Yeast SLI data is a valuable resource for mining therapeutically relevant interactions.
