< Back to previous page

Project

Applications of deep learning for cancer genomics

Since its selection as the method of the year in 2013, single-cell technologies have become mature enough to provide answers to complex research questions, including cancer genomics. With the growth of single-cell profiling technologies, thousands of genomic features can be measured simultaneously on tens or hundreds of thousands of individual cells in a single experiment. This allows us to characterize cell states in heterogeneous tumors at unprecedented depth. In addition, there has been a significant increase in data collected from single-cell profilings, resulting in computational challenges to process these massive and complicated datasets. To address these challenges, deep learning (DL) is a competitive alternative for single-cell analyses besides the traditional machine learning approaches. At the same time, the massive amounts of data create exciting opportunities to train DL algorithms for predictive modeling. Applications of such algorithms have been shown to classify patient samples, predict treatment outcomes, and (perhaps even more interestingly) provide mechanistic insight into the molecular processes underlying disease progression. We are now exploiting these methods to reconstruct the gene regulatory networks that convert healthy cell types into cancerous cells.

Date:1 Jan 2023 →  Today
Keywords:cancer, genomics, gene regulation, deep learning, AI
Disciplines:Computational transcriptomics and epigenomics, Single-cell data analysis, Development of bioinformatics software, tools and databases
Project type:PhD project