Walter Kolch - Cancer Signalling
Name: Prof. Walter Kolch
Job Title/Professional Qualifications: Prof., MD / Director, Systems Biology Ireland, Director, Precision Oncology Ireland
Research keywords: Precision oncology, systems biology, signal transduction
Current research projects: precision oncology of childhood cancers and melanoma; immunoncology of ovarian cancer; development of digital twins; multi-omics data integration
Contact details:
• Email: (opens in a new window)walter.kolch@ucd.ie
• Twitter: @sysbioire
• LinkedIn: https://www.linkedin.com/in/walter-kolch-147035115/
• ORCID iD: 0000-0001-5777-5016(opens in a new window)https://www.linkedin.com/in/walter-kolch-147035115/
• Web profile: (opens in a new window)https://people.ucd.ie/walter.kolch
Highlight Publication: Santra et al. (opens in a new window)An Integrated Global Analysis of Compartmentalized HRAS Signaling. Cell Reports. 2019 Mar 12;26(11):3100-3115.e7. doi: 10.1016/j.celrep.2019.02.038.
What is the significance of this publication? In this work, we have developed a flexible and universal framework for the integration of different types of 'omics' data. We now can generate many different types of omics data including genome sequencing, gene expression, protein expression and metabolite levels. However, we usually analyse them separately losing the natural context. We do this because data integration is very challenging and an unsolved problem. If we truly could put all data seamlessly together, we could deepen or understanding of how cancers arise and step-change our efforts to treat them. Our approach, called MiNeti (Mixed Network Integration), can integrate and analyse many data types. In this work, we have applied MiNeti to multi-omics data from cancer cell lines discovering important new functions of the RAS oncogene. Looking forward, MiNeti will allow us to gain very deep insights into the molecular pathogenesis of cancers. Importantly, MiNeti is a key step towards the construction of ‘digital twins’, which will allow us to simulate diseases and treatment responses in silico. Thus, we can safely try out different treatments on computer models in order to choose the best one for each patient.