Simplifying Delays in Systems Biology | Niklas Korsbo | Julia SciML | SIAM CSE 2021 скачать в хорошем качестве

Simplifying Delays in Systems Biology | Niklas Korsbo | Julia SciML | SIAM CSE 2021

Simplifying Delays in Systems Biology Niklas Korsbo, University of Cambridge Abstract: Most molecular interactions are usually ignored when modelling biochemical systems. This makes sense since the aim is often to represent a biological system with a model that is as simple as possible while still sufficiently faithful to the original system to be informative. However, while some interactions are so simple that their explicit inclusion in a model is unwarranted, the aggregation of many such interactions might not be. In a recently published paper, we take a look at the simplification of seemingly trivial multi-step pathways which tends to be simply ignored in models. We show how such multi-step pathways can cause signalling delays and how models that ignore them can be rendered unable to account for the dynamics of the modelled system. Instead, we propose an alternative assumption that allows for the reduction of the pathway into a gamma-distributed delay and we show how this is usually better able to capture the dynamics of the system. Getting delays right sometimes matters greatly for the quantitative accuracy, or even the qualitative behaviour, of a model. Gamma-distributed delays have long been used in fields such as electrical engineering and control theory but its use in pharmacology and systems biology is just beginning. Here, we probe its effectiveness, map its components to observables in biology, and provide a theoretical justification for its use. With this, we hope to make systems biology a bit simpler and a bit more accurate. For more information on the SciML Open Source Scientific Machine Learning Software Ecosystem, see https://sciml.ai/. For more info on the Julia Programming Language, follow us on Twitter:   / julialanguage   Contents 00:00 Speaker Presentation: Niklas Korsbo 00:48 Big Picture: A conceptual introduction to mathematical modeling for systems biology 02:40 Linear Pathways: Chains of events are ubiquitous in systems biology 03:15 Modeling challenge: How to effectively simplify linear pathways 05:34 Output Dynamics: Characterising the "sharpness" of the delay distribution 08:34 Alternative simplification: Fixing rates while tuning the pathway length 11:16 Better Simplification: From Erlang to Gamma distributed delay 12:21 Summary: What we've done and what we've obtained 15:24 Conclusion: Distributed delays are high-fidelity simplification of linear pathways 17:02 Acknowledgement and References S/O to https://github.com/pitmonticone for the video timestamps! Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/JuliaCommunity/You... Interested in improving the auto generated captions? Get involved here: https://github.com/JuliaCommunity/You...