Integrative computational and systems biology approaches for deciphering cellular regulomes that promote carcinogenesis

It is becoming easier to generate large cancer related datasets. However the methods to understand these and to extract meaningful, biologically relevant readouts from these data sets is underdeveloped. We are developing computational and statistical tools, methods and resources to integrate, mine, model and visualize these complex data-sets and convert them into a format that can advance our knowledge of cancer biology. This will enable a better comprehension of the circuitry involved in tumour formation, disease progression and metastasis, and can be applied to the development of targeted therapeutics.

Numerous genetic aberrations affecting a large number of gene products contribute to tumour development and the accepted hallmarks of cancer, often making therapies based on single targets ineffective. These aberrations and their consequences are propagated via transcriptional, translational and post-translational networks. The resultant information circuits can extend across cell and tissue types and also modulate the microenvironment. How to integrate and model this percolation of information and properties across different biological scales, in both time and space is accepted as a major challenge to fully understanding the aetiology, progression and dynamics of tumorigenesis. Novel computational methods, tools and resources are essential to address this challenge.
Transcriptomic and copy number assessment studies suggest that specific sets of key pathways that regulate cellular phenotypes are pathologically altered in cancer. However, pathway centric approaches are not completely effective because pathways don’t operate in isolation; instead, multiple interacting pathways form a complex web of networks. Methods to decipher cellular information flows (whether they are signalling, metabolic or regulatory) can shed light on the underlying cellular complexity and identify systems level properties. The objective of this research programme is therefore to develop a network perspective of tumorigenesis using a mixture of both integrative data driven as well as computational modelling approaches.
The programme harnesses data science approaches coupled with computational and systems biology methodologies to develop novel methods, tools and resources for systems level biomedical data-integration. It addresses these objectives by 1) modelling multi-scale cellular regulomes involved in tumorigenesis, 2) developing these model into knowledge discovery systems to assist in understanding of cancer datasets, and 3) Using these resources to elucidate phenotypic impact and identify therapeutic intervention points of these regulomes for further experimental evaluation.