From a methods perspective, our primary interest is in the development of the data science and machine learning expertise for the analysis of data geneated by single cell profiling assays across different modalities and in different contexts. You can read about these here. We have a long history of working in breast cancer informatics and genomics, and continue to develop analytic methodology in this direction. There are a number of on-going computational projects, many of which are rooted in deep learning.

  1. Analytical techniques to capture cellular response to chemical perturbagens. This work has primarly been done in the context of our fungal C. albicans projects. The work is based on extensions to GAN and VAEs entities from deep learning.

  2. Computational vision tools for mycology. This work is primarily based on extensions to FASTER R-CNN models.

  3. Methods to exploit multi-modal single cell data. This is ubiquitous throughout our projects, and relies heavily on basic statistics and visualization through deep representations.

  4. The development of gene and gene product signatures related to clincial endpoints primarily in breast cancer. See for example our CIHR-funded DCIS project.

  5. The inference and representation of biological networks. See for example this.

This work has been developed with different members of our team.