Poster: AI-driven identification and validation of novel synthetic lethal gene pairs through deep mining of cancer dependency data

Download the poster

Conclusions

• Evariste have performed in silico modelling of cancer dependency data to identify the next generation of druggable synthetic lethal gene pairs
• We have prioritised gene pairs with a rational mechanism linking the target and biomarker, and with patient populations with high levels of unmet need
• These pairs reflect both pan-cancer and highly lineage specific relationships
• We have validated >10 of these relationships across multiple cell lines to give high confidence in the validity of the novel pairs
• For one gene pair with a kinase as the druggable node we have generated novel inhibitors with best-in-class selectivity over the SL partner and properties suitable for in vivo dosing and validated the specific targeting of biomarker low cell lines