Sven Nelander’s research on the systems biology of neural cancers

Precise targeting of cellular networks in brain tumor stem cells

Our lab is developing novel strategies to enable efficient network-based drug development to target cancer stem cells (CSCs). For this, we work with partners to establish a unique Uppsala based biobank of more than 150 patient-derived CSC cultures, from patients with brain cancer (glioblastoma).

To enable a systems biological analysis, each cell line is systematically characterised at multiple levels. Unlike traditional biobank studies, our characterisation includes both functional and molecular data, ranging from mutations, to epigenomics to comprehensive knockdown screening information. We then construct computational models that aim to increase our understanding of CSC biology, including:

  1. Which are the mechanisms that drive key phenotypes of the CSCs, like tumour initiation capability?
  2. Which are the mechanisms that make mediate functional heterogeneity, e.g. differences in drug response between two patients?
  3. How can we optimally intervene optimally to suppress disease progression or prevent recurrence after surgery?

An important unique aspect of this study is the integration between a state of the art biobank with computational modelling of extensive data. The effort thus has potential to unravel new therapies, patient prognostics and biomarkers. Our effort is highly inter-disciplinary and involves collaborations with the IGP neurooncology groups, SciLifeLab platforms as well as international partners.

Big data integrative models of cancer

The ongoing efforts worldwide to develop cancer therapies are increasingly dependent on accurate data analytics. My lab develops new methods computational methods and mathematical models that will help researchers to interpret complex cancer data sets. Key challenges that we are addressing are:

  1. How can multiple sets of cancer information be computationally integrated into models that help us understand cancer mechanisms?
  2. Can we predict strategies to protect non-cancerous tissue from side effects of cancer therapies?
  3. How do we best design combinatorial interventions against cancer cells?

Addressing these multi-faceted questions, we combine both unqiue, in-house data sources, as well as multiple layers of public data. A key component of the work is also to make our results available as tools and packages that can be used by cancer researchers. One recent example is (NAR 2015). 


Cancerlandscapes, a new data mining tool for cancer researchers.
Cancerlandscapes is a new SciLifeLab web resource for data mining of different cancer diseases. The figure depicts the user interface, showing regulatory network of brain tumors.