Machine Learning Research

The group has a long history of developing large-scale learning methods for learning to classify sequences, predicting annotations of very long strings such as genomes and solving large optimization problems using iterative methods.

Computational Transcriptomics

With the advent of high throughput sequencing technologies, genomics and transcriptomics experienced a challenging renewal. Notably, RNA-Sequencing enhanced transcriptome analysis and opened great opportunities for the gene discovery and the identification of alternative transcripts. We made numerous contributions to the field in tackling problems such as isoform identification and quantification, differential expression analysis and the identification of alternative splicing events.

Molecular Data Science

RNA plays a central role during the whole lifetime of a cell. It serves not only as a messenger towards protein expression but also as a regulator of gene expression, transcript modification or translational processes. To elucidate the relevance of sequence and structural features of RNA for the outcome of these processes and to ultimately understand the complex organization of the cell’s regulatory system, sophisticated computational methods need to be developed and applied. High throughput measurements and large sample sizes, that are both essential for these kinds of studies, raise the demand for feasible and easily scalable computational solutions.

Clinical Data Analysis

The group develops innovative methods for the analysis of electronic health records (EHR) and pathology slides with the objective of automatically summarizing patient states over time. The overall aim is to build elements of decision support systems that utilizes EHR, image and genomic information pieces to provide comprehensive, automatic suggestions for treatments, to assign patients to clinical trials and to assist in the design of new trials.