Jesper Tegner

Brief Biography


Jesper Tegner is a Professor in Bioscience and Computer Science at KAUST (2016) and an Adjunct Chaired Strategic Professor of Computational Medicine at Karolinska Institutet. He obtained the rank of chaired full professor 4.5 years after his PhD. He holds three separate undergraduate degrees with majors in MedSchool, Mathematics, and Philosophy respectively. Next he did two years postgraduate education in pure and applied mathematics, while doing an experimental Ph.D./M.D. degree in Medicine. He was a Visiting Scientist as a Wennergren and Alfred P Sloan Fellow (USA, 1998-2001) while holding a position as an Assistant Professor of Computer Science 1998-2002 at Royal Institute of Technology.

He is an ERC co-investigator (consolidator) and is ranked as outstanding (highest distinction) at Karolinska Institutet. He is an acting Section Editor on Clinical and Translational Systems Biology in Current Opinion on Systems Biology, Senior Editor in Progress in Preventive Medicine, and serves on the editorial Boards for BMC Systems Biology and Neurology: Neuroinflammation & Neurodegeneration. He is a fellow in the European society for Preventive Medicine.

As a founding director of the Unit of Computational Medicine, an interdisciplinary team of 30 researchers, his published work (close to 200 publications) includes Science 2005, Journal of Machine Learning Research 2007, 2009, IEEE Systems Biology 2009a, 2009b, NAR 2012, Nature Genetics 2009, PNAS 2002, 2003, 2004, 2007, 2009, 2015, Trends in Genetics 2006, Immunology and Cell Biology 2016, Epigenetics 2013, 2014, Genome Medicine 2009, 2016, Nature Communications 2015, Cell 2010, Nature Methods 2016, Scientific Reports 2015a, 2015b, 2016. Seminars in Developmental and Cell Biology 2015a, 2015b. He has founded two BioIT companies (2003, 2008), where one became the winner (2005) of a national award for being the most promising company of the year.

His team is interested in decoding molecular circuits of living systems with special reference to collective dynamics of single cells and populations thereof. His team develops and applies experiments, theory, and computational techniques to uncover regulatory circuits in cells. Some

discoveries include specifics of regulatory circuits in immune cells and design of computational network and modeling techniques. Using these techniques in combination with carefully designed experiments his team has elucidated novel mechanisms in the generation of atherosclerosis, stratification of Rheumatoid patients, clinically decision systems targeting COPD, integrative analysis of Parkinsons´ and Alzheimer´s disease, and biomarkers for different stages of multiple sclerosis.

By practicing an integrative experimental and computational approach, guided by theory, investigating adaptive molecular circuits in living systems, the team, believes that such circuits could hold secrets to new properties beyond what is readily apparent from the fundamental equations of matter. Examples include principles of natural adaptive computation, essentially protecting living organisms from lapse into atomic chaos. Such algorithms, inspired from living systems, could be harnessed in the design of novel technologies, innovations, medical applications, and ultimately in crafting general reasoning systems.

From Data-driven Reconstruction of Cellular Circuits to Generative Causal Models

________________________________________________________________________________________________________

Our ability to discover statistical patterns has developed dramatically with the systematic generation of large-scale biology data sets. Yet, numerous questions ranging from biological understanding to therapeutic decisions require mechanistic insight into the underlying generative processes. The need for predictive interventional knowledge beyond observational descriptions increases with big heterogeneous data in order to find and prioritize needles in the haystack.

Mathematical modelling and computer simulations are powerful tools but our ability to identify, formulate and analyse complex non-linear models is severely hampered in the absence of first principles, and when knowledge or data are incomplete. Here I will discuss this fundamental divide in the context of our recent experimental and bioinformatics analysis of cellular circuits governing reprogramming of human Tcells into specific Tregulatory cells, critical in the control of the adaptive immune response, implicated in cancer, inflammatory, and degenerative diseases,

A data-driven integrative experimental and bioinformatics time-resolved network analysis identified several known and novel molecules participating to the iTreg generation, as well as known regulators such as the master controller, FOXP3. Several of the candidates were experimentally validated. Whereas such global network analysis yields locally validated new biological insights of putative clinical value we still face the challenge of a large-scale causal deconstruction. I will therefore also discuss our ongoing work on how to mitigate this fundamental gap between data and generative models using an ensemble modelling approach.

In conclusion, while molecular network analysis is readily applicable across cellular systems there are open challenges on how to reach quantitative predictive generative causal models from such large-scale networks.