Entering edit mode
8.3 years ago
jcbtbs
•
0
I went to a talk by Dr. Jonathan Karr of Karr Lab yesterday and was completely blown away by the possible implication of his research. The paper that his talk was based on is called A Whole-Cell Computational Model Predicts Phenotype from Genotype, and I'm just wondering what you think is possible with this modeling technique? Assuming we can make the necessary strides in improving this technology to successfully model an entire human cell, what do you think are some of the most exciting possibilities?
Thanks for the response, Jean-Karim. Would an example of such a hypothesis be: "Does Jane's skin cell produce melanoma faster than Jon's skin cell when they are both experience long exposure to sunlight?" or "What happens if we give patient X 100mg of some medicine instead of 500mg?". Are these the types of questions that can be answered more easily with enhanced whole-cell modeling?
This would be possible if/when you could individualize/personalize the model. Before this we would most likely have a model of a generic cell where you could ask what happens if you introduce some perturbation e.g. you could compare the effects of concentration x vs concentration y of a compound. Note that there are other modelling approaches than the one used in the paper you mentioned that could also have the potential to get to the level of virtual experiments.
Cool, thanks for the input. Can you provide some examples of other modeling approaches that have the potential to get to the level of virtual experiments?
There are approaches based on extensions of Boolean networks e.g. this paper, agent-based models (here). Specific systems can also be modeled using physical models e.g. particle-based models (here) or active gels (here).
Thank you, Jean-Karim. I appreciate it.