Dr. Johann de Jong

Developing methods grounded in machine learning (specifically deep learning) and statistics, for patient stratification and drug response prediction from clinical and multi-omics data.

Integrating multi-level biological data via computational and statistical means, and driving the development of new computational methodologies for this purpose.

Working on 1) Multiscale normalization of genome-wide sequencing data using a kernel convolution-based method, 2) Integrating large genomics and epigenomics datasets for studying (i) the efficiency of the CRISPR-Cas9 system for genome engineering and (ii) the chromatin position effect across multiple promoters.

Using techniques mainly derived from machine learning and statistics (HMMs, Bayesian networks, kernel convolution, etc.) to analyze very large genomics and epigenomics datasets, and developing algorithmic tools, mainly focusing on the applications of retroviruses and transposons in cancer research and gene regulation.

Mathematical modeling, complex systems, high-performance and distributed computing, artificial intelligence, scientific computing, scientific visualization, computational biology.

Pattern recognition and machine learning, NGS data analysis, gene regulation networks