Our research focus on advanced data analytics in life science and use modern data science and engineering tools to develop computational models to understand, simulate and predict behaviour of complex biological systems.
We combine multivariate analysis, deep learning and engineering modeling tools and approaches in open and strategic collaborations with academic leaders and industry partners.
Multivariate data analysis (MVA) has always had a strong focus in our group. Here, developments include OPLS and multi-block methods (O2PLS, OnPLS, JUMBA) that are particularly established in “omics” disciplines, especially in metabolomics. Such methodologies have consistently demonstrated improved biological interpretation and identification of correlative biomarkers, ranging from large scale biology studies to smaller clinical studies. We continue active research in the area of multi-block analysis and data integration.
Design of experiments is another focus area that is crucial to generate high-quality and representative data in life science, including sample selection and experimental workflows in life science where we continue an active line of research.
In systems biology research, we target the integration of biological information across multiple phenotypic and “omics” platforms. Here, we are continuously developing multivariate tools and strategies to analyse and integrate omics data for enhancing the interpretation of biological data. In recent years, we’ve also added modern data science tools including deep learning, and mechanistic modelling tools and approaches into our computational workflows.
Recently, we are also conducting projects in analysis of large-scale cell imaging data (1 billion cells) within the JUMP consortia (PI: Broad institute/Anne Carpenter) focused on creating a global phenotypic atlas for drug discovery and drug development, this is also in collaboration with biopharmaceutical industry. This includes high-content, high throughput image-based assays used to assess the effects of molecules and genetic perturbations on cells using machine learning approaches.