We are making use of machine learning and network pharmacology approaches to integrate single-cell omics profiles with drug-target interaction networks to identify patient-specific, cancer-selective and synergistic drug combinations that effectively co-inhibit multiple cancer driving sub-clones and other escape routes of cancer cells.
We are using lineage tracing approaches to dynamically trace each cell’s progeny and to identify key molecular switches that determine cancer microevolution paths during disease progression and therapy resistance. Medical aim is to develop novel treatment approaches for targeting tumor-microenvironment interactions and therapy resistance.
We are implementing computationally efficient and robust statistical and machine learning models for mining combinations of molecular and clinical features most predictive of patients' outcomes, such as differences in disease risk or treatment responses, which may eventually provide predictive signatures for patient stratification.
Drug repurposing is an attractive approach because of its potential to speed-up and de-risk the drug development process, and to provide treatments for unmet medical needs. We are developing AI-based methods, open-access databases, and software tools that support FAIR principles to repurpose drugs for cancers and other indications.