The Systems Biology of COVID-19 and the SARS-CoV-2 virus. The class will build a foundation that includes the emergence of complexity, simple biological subsystems, their reductionist and equivalent toy and organ-chip models, and the measurements required to specify model architecture and parameters. Applications to biology, physiology, medicine, chemical and biological defense, pharmacology, drug discovery, and toxicology. UGrad: PHYS 240 01 and BME 290B; Grad: PHYS 326 and BME 395C.
Saturday, November 12, 2016
SysBio16 Asgn_24B_Class_24_Article_20_2016_11_17
CN-Bio QSP: Read-Scan Article 20 J Yu et al., Quantitative systems
pharmacology approaches applied to microphysiological systems (MPS):
Data interpretation and multi-MPS integration, CPT-PSP (2015). Reading
team - read carefully; Class - read abstract, captions, conclusions so
you know what's going on and can ask questions of our guests. Post a
PCRC on the blog.
Subscribe to:
Post Comments (Atom)
0. KNEW:
ReplyDeleteBased on previous articles and class discussion I was familiar with Microphysical Systems (MPSs) and why they are being emphasized for in vitro analysis. I knew that data acquisition from multi-MPS platform design is a goal of the devices, and that the scale of these systems can be as small as 1 inch by 1 inch silicon chips.
1. LEARNED:
Figure 1 shows the concentration of total hydrocortisone (HC) from the traditional pharmacokinetic (PK) model as well as the unbound HC concentration from the mechanistic PK model. The charts include 2 different time frames and 2 different concentrations of human serum albumin (HSA). Table 1 showed the mean metabolism rate constant (k) for the two models, suggesting that k is only dependent on HSA concentration for the non-mechanistic model. Figure 2 shows the inflammatory response of the liver/immune MPS with repeated doses of LPS. This figure shows the desensitization to subsequent doses of LPS, behaving as an adapted signal response. Figure 4 gives an example of a multi-MPS platform. Figure 5 shows simulations of this multi(4)-MPS platform. This figure investigates the time for the MPS and mixing chamber to become well-mixed following administration, exposure, and elimination of a drug. Figure 6 shows the concentration of the drug over time in each MPS as well as the mixing chamber. The importance of the selection of parameters for any given experiment are shown, suggesting trade-offs must occur.
2. PRESSING QUESTIONS:
Do the examples in figure 2 show a "sniffer" response? The graphs show a decreased response of normalized TNF-alpha and normalized IL-6 over time, but the dosing only occurred every 48 hours and was not constant.
3. PRESENTATION TOPIC:
How this paper relates to previous topics that have been discussed in class.
4. THOUGHTS:
The discussion brought up several good points regarding the multi-MPS platform, suggesting that the complexity should include mechanistic detail without including too many parameters. Good paper that relates to previous class discussions.
0. KNEW
ReplyDeleteI was familiar with microphysiological systems technology along with pharmacokinetic and pharmacodynamic models from previous classes and papers. I knew that PK/PD models address how the drug affects/is affected by the body in regards to absorption, distribution, metabolism, and excretion.
1. LEARNED
The scope of traditional PK/PD models is specific to in vivo models, and is too limited for use in in vitro systems and data interpretation of data from the MPS operation. Instead, computational models w/ greater mechanistic detail with physical and biological dynamics would be used to design MPSs, select experimental conditions, analyze and predict outcomes.
It seems a system of MPSs are referred to as 'interactomes'.
Experiments comparing traditional PK models to the four-MPS interactome model show that the interacome model was more robust. For example, it's shown that the interacome model's k(metabolism, unbound) was constant with changing [HSA], while the traditional PK model had an nonconstant k. The second experiment with inflammatory response to LPS stimulation, complex dynamics prevent a traditional PK/PD model from explaining the results where the interactome model could make key observations. Due to experimental restrictions, it's generally infeasible to mimic in vivo processes precisely - computational models are needed to translate the in vitro data to in vivo.
2. QUESTIONS
How well does this model scale to humans? I've asked this question in the TEDx presentation on OoC, but it seems to me that there will always be parameters not considered when downscaling to the MPS-scale.
3. PRESENTATION
A short presentation on transformations of a drug as it travels throughout the body up to excretion.
4. THOUGHTS
I'm seeing a large emphasis on 'data-driven design' both in this paper and in other papers such as those about machine learning; I'm a little confused on what design would not be data-driven.