Cancer biology and treatment involves complex, dynamic interactions between cancer cells, the tumour microenvironment, and therapeutic molecules. Quantitative approaches combining mechanistic disease modelling and computational strategies are increasingly leveraged to rationalize pre-clinical and clinical studies, and to establish effective treatment strategies. In this way, mathematical approaches lay the foundation for computational “virtual laboratories” that offer fully controlled, and non-invasive conditions in which we can investigate emergent clinical behaviours and interrogate new therapeutic strategies. As an introduction to such virtual laboratories, this workshop will provide an overview of techniques used in computational oncology, with a focus on in silico clinical trials and agent-based models (ABMs). Virtual (or in silico) clinical trials are useful computational platforms that help distinguish mechanisms of therapeutic successes and failures, stratify patient risk classes based on an individual’s physiology, and optimize drug-specific parameters. In these platforms, in silico patients are generated by drawing from distributions of possible patient characteristics and used to form virtual clinical trials, in which new treatment strategies can be evaluated prior to human trials. Data fitting and optimisation techniques are cornerstones of this computational platform and are used to generate realistic virtual patients and evaluate individualised therapies. ABMs are a computational formalism that describes the way individual agents (e.g. cancer cells) interact through probability distributions based on defined characteristics that have contributed significant insights into cancer biology at the intra-patient tissue level. In oncology, this technique has been applied to model spatial tumour formation, tumour cell heterogeneity, and the dynamics of treatment in the tumour microenvironment. Modelling individual cells as agents allows for direct translation of biological observation into simulation rules and, like virtual clinical trials, the investigation of new hypotheses and treatment strategies.

In particular, this workshop will address:

• the optimization of parameter ranges to generate virtual patients or treatment schedules using a variety of techniques, including simulated annealing, least-squares nonlinear optimisation, gradient-based descent, and genetic algorithms.

• the translation between ABMs and PDEs

• how to code heterogenous tumour environments into an ABM using an open-source software known as PhysiCell

The workshop is free but the participants need to register by clicking here:

https://www.eventbrite.com/e/computational-modelling-to-study-cancer-bio...

The zoom link of the workshop will be send to registrants emails few days before the workshop.