Systematic discovery of mutation-specific synthetic lethals by mining pan-cancer human primary tumor data

Area of Science:

• Computational biology
• Genomics
• Cancer research

Background:

• Synthetic lethality (SL) describes conditions where mutations in two genes are lethal together but not individually.
• Identifying SL partners for cancer mutations is crucial for targeted therapies but challenging with current methods.
• Existing cell line-based approaches face limitations in identifying clinically relevant SL interactions.

Purpose of the Study:

• To develop a novel computational algorithm, MiSL (Mining Synthetic Lethals), for identifying mutation-specific SL partners from pan-cancer tumor data.
• To predict SL partners for a wide range of mutations across various cancer types.
• To discover genetic biomarkers for predicting drug sensitivity in cancer patients.

Main Methods:

• Developed the MiSL algorithm to analyze large-scale human primary tumor mutation data.
• Applied MiSL to 12 different cancer types, predicting numerous SL partners for thousands of mutations.
• Validated MiSL predictions through comparisons with existing functional screens and experimental models.

Main Results:

• MiSL predicted 145,891 SL partners for 3,120 mutations across 12 cancers, including known SL interactions.
• Predictions showed enrichment for true SLs in multiple cancer types when compared to functional screens.
• Successfully validated a specific SL interaction (IDH1 mutation and ACACA) in leukemia using gene targeting and patient-derived xenografts.

Conclusions:

• MiSL effectively identifies mutation-specific synthetic lethal partners from pan-cancer genomic data.
• The algorithm accelerates the discovery of novel therapeutic targets and genetic biomarkers for precision oncology.
• MiSL demonstrates significant potential for advancing targeted cancer therapy development and patient stratification.