Tissue histopathology gold-standard for disease diagnosis/prognosis. But, it is difficult to relate molecular changes/pathway to tissue-level outcomes, or to relate tissue effects to molecular perturbations. We need to describe the relevant biology to simulate dynamic dose/time-dependent phenotypes. Their goal is really to understand the molecular dymanics.
The goal of their work is to develop better in vitro systems and predict pathological outcomes. However, molecular perturbations/pathways alone are insufficient to understand functional alterations in tissues/organs. As an example of these tissue-level changes, take cell aterations based on toxicity: swelling, steatosis, macroves or microves, necrosis, hyperplasia, and carcinoma are just a few examples. If you look at the gene expression profile for these alterations, it’s not easy to parse out which change is the one that’s occurring.
In order to link molecular pathways to phenotypes, they’re working on using virtual tissues, which are multi-cell models. They’re not trying to build a holistic model of the entire cell: instead, they’re focusing on cell behaviours that define key molecular pathways and define the key cell-cell interactions.
He then mentioned the V-Tissues conference that happened in April 2009 that went well, and their website for the conference. At this conference, they had transitional goals (bridging gaps). Some examples of tissue-level models (liver toxicity and cancer, immune response, cardiac modelling and more) and experimental and computational needs were given of what went on at the workshop. There were also breakout discussions which covered computational requirements (modelling and simulation frameworks and formalisms, and cell/tissue-level knowledge integration via ontologies to capture meaning and reduce ambiguity) and data requirements (organotypic cultures and histomorphometry).
They need to define use-cases. (Personal note: use cases and case studies are very important!)
The knowledge representation issues they’re dealing with include: key events in cell response, cell-cell and cell-ecm interactions, and microcirculation. They’re also interested in cell changes, such as in cases of chronic toxicity in cancer. They are very histopathologically focused.
They’re integrating this information using the V-Liver architecture, which goes from assays to v-Liver Knowledgebase, v-Liver simulator, and from there to outcomes. He mentions that some of these steps would involve the use of the semantic web, but didn’t go into details. They have a key need for fundamental cell behaviours (cellular outcomes, and key intracellular processes and outcomes.
Please note that this post is merely my notes on the presentation. They are not guaranteed to be correct, and unless explicitly stated are not my opinions. They do not reflect the opinions of my employers. Any errors you can happily assume to be mine and no-one else’s. I’m happy to correct any errors you may spot – just let me know!