We design algorithms to understand gene regulation and DNA damage and repair, to characterize disruptions of these processes in disease, and to discover therapeutic targets.
- Machine learning
- Data mining
- combinatorial algorithms
- n-way clustering
- time course analysis
- domain-driven data integration
- Transcriptional regulation
- Epigenetic regulation
- DNA damage and repair
- Gene expression
- TF-DNA binding
- Gene disruption and editing readouts (CRISPR + sequencing)
- Gene essentiality
- Mutation patterns and signatures
Focus on disruption of relationships and pathways rather than single functional elements (i.e. genes).
- Major regulation of disrupted pathways.
- Mechanisms involved in drug resistance.
- Gene essentiality and synthetic lethality relationships.
- DNA damage and repair.
Some of our current projects.
Learning algorithms for cancer therapeutic target prediction (starting in 2020)
machine learning, essentiality (CRISPR, RNAi), synthetic lethality, molecular profiles
Modelling of single-cell heterogeneity in proton therapy response (starting in 2020)
machine learning, proton therapy, head and neck cancer, single-cell, sequencing, optochemistry, imaging
Uncovering key regulators and pathways downstream the androgen response in prostate cancer
gene expression, time course, TF binding, prostate cancer, cell line, knockdown
Temporal 3-way clustering algorithm to mine local response patterns over time across samples (e.g. patients, single-cells)
algorithms, string matching, combinatorial pattern mining, suffix trees, itemset mining, time series, biclustering
Some of our recently completed and published research.