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.
Friday, October 11, 2019
SysBio19 Asgn_5_Class__Woo_article in Networks Folder_2019_10_17 - Due, Tuesday, Oct 15!
Read carefully Woo article in Networks Folder: Elucidating Compound
Mechanism of Action byNetwork Perturbation Analysis. Answer the
questions: 1) What did you already know? 2) What did you learn? 3) What
is your most pressing question that you would
like Blake, Cody, and Tejas to explain in class? 4) How does this
relate to your project?
Subscribe to:
Post Comments (Atom)
1) What did you already know?
ReplyDeleteI knew that there were methods of creating the interactions but didn't know DeMAND.
2) What did you learn?
I had never heard of DeMAND and how its useful to find out interactions in these systems.
3) What is your most pressing question that you would like Blake, Cody, and Tejas to explain in class?
Can you explain more about how DeMAND is done?
4) How does this relate to your project?
This is another tool for being able to model neurons especially helpful in trying to build the boolean network.
1) What did you already know?
ReplyDelete* I knew the general mechanisms of action of small molecules and how those mechanisms of action are deduced based on the molecules' affinity for specific membrane-bound signalling proteins.
2) What did you learn?
* The DeMAND algorithm used perturbation data from previous studies to elucidate the mechanism of action and pinpoint specific effectors of altered cellular function. It was interesting to see an algorithm integrate such data into useful predictions.
3) What is your most pressing question that you would like Blake, Cody, and Tejas to explain in class?
* I cannot get a solid grasp on how the DeMAND algorithm actually goes from data to prediction of MoA. I would like a case example walking me through how, for a specific compound, the DeMAND algorithm operates and how it makes the predictions it does.
4) How does this relate to your project?
* This algorithm is extremely interesting to me because of its potential to identify key players in specific functions of cells and how those players may be manipulated for cellular dedifferentiation. I also think that it is useful to have a more robust grasp of the mechanisms active in the cell for future questions.
1) Small molecules mechanisms can be determined based on affinity to proteins, I knew of other machine learning methods of creating networks
ReplyDelete2) Perturbation data can be find mechanisms and determine specific effectors when cell function changes
3) Specific algorithm, math behind prediction method
4) A similar algorithm can be used to elucidate relationships between phenotypes, genotypes, and environmental factors. If perturbations are defined in these data sets, networks for disease (or other) mechanism of action can start to be created.
1) I already knew that you can analyze small molecules through their interactions with proteins. I know how to make networks of interactions from RNASeq data but not from small molecule data.
ReplyDelete2) I learned about their DeMAND algorithm and how it is much more effective at identifying predictions than a simple t-test.
3) I don't understand the math behind the algorithm. I also don't understand their validation using siRNA.
4) We could try to find data on how perturbations can effect the current pathways that researchers can use to take stromal cells to iPSCs.
1) Knew: I'm familiar with KL divergence as a measure of information loss between two probability density functions, as well as using statistical tests for differential expression
ReplyDelete2) Learned: combining expression correlation/difference with interaction networks can be used to elucidate mechanisms of perturbation within a system.
3) Question: How could you co-opt this for novel GRN discovery? I'm guessing that you can use a similar approach to compare network interactions in different knockout or growth conditions and maybe learn more about the networks themselves.
4) Project: Could be good experiments for directed cell differentiation, comparing different conditions. Might elucidate previously uncharacterized mechanisms/genes involved in cell differentiation/fate decisions.
1) What did you already know?
ReplyDeleteI already know that small molecules have very significant impacts physiologically and that determining mechanisms of action are incredibly important, but very labor intensive to accomplish. I also somewhat knew that machine learning application here.
2) What did you learn?
I hadn't really thought about the algorithm that would be behind understanding mechanisms of actions in a far more efficient way. I need to go into more depth regarding the algorithm in this paper.
3) What is your most pressing question that you would like Blake, Cody, and Tejas to explain in class?
n/a
4) How does this relate to your project?
This is very impactful in so many ways. From pharma companies who currently mass tests small molecules to see if there is any effects, it can make that process more efficiently. In our case, if we are looking at cell differentiation, understanding the mechanism behind that could be useful by understanding effects of certain small molecules in that process.
1) What did you already know?
ReplyDeleteBasics of MOA and the assays used to determine MOA.
2) What did you learn?
The Demand algorithm uses network analytics to predict MOAs for compounds of interest. This algorithm has been validated and can elucidate differences in compounds with similar MOAs.
3) What is your most pressing question that you would like Blake, Cody, and Tejas to explain in class?
How are the initial networks determined and integrated?
How does the needed data for determining the p-values scale with network size? At what point is a network too large to be feasibly analyzed?
4) How does this relate to your project?
This algorithm can be used to predict potential affectors of my DA neuron networks.
1) What did you already know?
ReplyDeleteDifficulties in elucidating complete MOA, especially off-target effects
2) What did you learn?
DeMAND is an algorithm that can elucidate MOA of a drug or compound through comparison of GEPs of a control and a compound perturbed group in terms of assessing dysregulation of gene-gene interactions
3) What is your most pressing question that you would like Blake, Cody, and Tejas to explain in class?
I do not understand the statistical methods including how they are calculating the probability densities
4) How does this relate to your project?
If we had a gene regulatory network to look at for dedifferentiation, we could predict MOA of compounds that are known to induce dedifferentiation in order to identify key regulatory genes and increase efficiency
1) What did you already know?
ReplyDeleteNot much. We've talked about the effect of small molecules on cell behavior in the past, but I hadn't appreciated their significance in cell function and the importance of their MoA.
2) What did you learn?
The significance of MoA in determining potential toxicity in drugs, in addition to the difficulty in determining the MoA of small-molecule compounds.
The algorithm DeMand for determining MoA by monitoring which genes are dysregulated by certain compounds. It was able to deduce these networks by comparing gene regulation/dysregulation data from a control group to that of a compound perturbed group.
3) Questions for the team?
How has this method been received in comparison to the others they listed? They mention that other methods are time/slash labor intensive, so this seems like a natural replacement. Is it gaining traction, or has not found much use in other studies.
4) How does this relate to your project?
The cardiac team is currently interested in how certain signaling proteins - TGFB, for example - effect cardiomyocyte/fibroblast (de)differentiation. It would be useful to know if this technique could be applied to proteins of this size.
If not, applying it to various other compounds known to cause cardiac cell differentiation could give us a better grasp of potential side effects i.e. toxicity and other pathways that are/could be activated by certain compounds.