𝕏

Previous Model vs New Model

Message boards : Science : Previous Model vs New Model
Message board moderation

To post messages, you must log in.

AuthorMessage
Yavanius
Avatar

Send message
Joined: 20 May 15
Posts: 50
Credit: 374,843
RAC: 428
Message 1676 - Posted: 8 Jul 2022, 5:47:28 UTC

Jesús,

Can you simplify the differences (layman terms) the difference between the new model and the previous model?

At a glance, it seems the newer model allows a greater range of options in your experiments than the previous model. it's a bit late, so the mind isn't quite grasping the difference between the models. :)

TIA,

Yav
ID: 1676 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Jesús Carro
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 18 Mar 15
Posts: 275
Credit: 1,956,954
RAC: 22,420
Message 1687 - Posted: 11 Jul 2022, 15:40:07 UTC - in response to Message 1676.  

The difference between the previous model and the new one is very small. The structure of the model is the same. The model is composed by currents that go through the cardiac cell membrane. We have modified the parameters of two currents to solve some issues we found (same mathematical expression but with different constants). Now we need to compensate this modifications to get the whole model working well. This is the objective of the simulations that you are running.

We have also added more parameters to test more experimental protocols. Now we are testing 45 protocols that permit us to check more than 100 markers... Many more information to try to fit the model to the experimental data.

We have also other markers that we are not using now. We will use them for the validation of the model... in the next stages.

I don't know if this explanation is simple enough or if it is what you were expecting. If not, I will try again 😅.

Best,
Jesús.

==============================

La diferencia entre el modelo anterior y el nuevo es muy pequeña. La estructura del modelo es la misma. El modelo está compuesto por corrientes que atraviesan la membrana celular cardíaca. Hemos modificado los parámetros de dos corrientes para resolver algunos problemas que encontramos (misma expresión matemática pero con diferentes constantes). Ahora necesitamos compensar estas modificaciones para que todo el modelo funcione bien. Este es el objetivo de las simulaciones que estáis ejecutando.

También hemos agregado más parámetros para probar más protocolos experimentales. Ahora estamos probando 45 protocolos que nos permiten comprobar más de 100 marcadores... Mucha más información para intentar ajustar el modelo a los datos experimentales.

También tenemos otros marcadores que no estamos usando ahora. Los utilizaremos para la validación del modelo... en las próximas etapas.

No sé si esta explicación es lo suficientemente simple o si es lo que esperabas. Si no, lo intentaré de nuevo 😅.

Un saludo,
Jesús.
Jesús Carro
Universidad San Jorge
@InSilicoHeart
ID: 1687 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Yavanius
Avatar

Send message
Joined: 20 May 15
Posts: 50
Credit: 374,843
RAC: 428
Message 1690 - Posted: 12 Jul 2022, 4:19:43 UTC

Assuming you can continue to receive funding, do you have an idea of how long DENIS models will run? Or does the nature of the project allow to be open-ended, that is, is there a finite set of parameters or is there an endless amount of models that can be run?
ID: 1690 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[VENETO] boboviz

Send message
Joined: 9 Apr 15
Posts: 171
Credit: 1,483,869
RAC: 1,039
Message 1693 - Posted: 12 Jul 2022, 7:29:12 UTC - in response to Message 1690.  

Assuming you can continue to receive funding, do you have an idea of how long DENIS models will run? Or does the nature of the project allow to be open-ended, that is, is there a finite set of parameters or is there an endless amount of models that can be run?


Good point.
And, another question: how our simulations are used? In vitro tests?
ID: 1693 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Jesús Carro
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 18 Mar 15
Posts: 275
Credit: 1,956,954
RAC: 22,420
Message 1701 - Posted: 13 Jul 2022, 15:32:37 UTC

No, I'm not sure for how long. Now we have simulations for months (probably), but this is not an open-ended. However, as long as we continue to have ideas in mind, we will continue to run things here.

If for a period of time we have to stop because there is nothing to simulate, we will notify you.

On the other hand, I think I don't understand the question of the invitro tests. Your simulations are used in the computer to get the best parameters, and later they will be compared with results of experiments. We are not going to carry the experiments, they are done and the results are published in scientific journals.
Jesús Carro
Universidad San Jorge
@InSilicoHeart
ID: 1701 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[VENETO] boboviz

Send message
Joined: 9 Apr 15
Posts: 171
Credit: 1,483,869
RAC: 1,039
Message 1711 - Posted: 14 Jul 2022, 12:17:42 UTC - in response to Message 1701.  

On the other hand, I think I don't understand the question of the invitro tests. Your simulations are used in the computer to get the best parameters, and later they will be compared with results of experiments. We are not going to carry the experiments, they are done and the results are published in scientific journals.


I don't understand.
We are simulating the response of heart's cells to some chemical compounds (or their combinations). Is it right?
So, who make the "in vitro" (or, later, "in vivo") experiments to validate our simulations? Are not you?
ID: 1711 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Jesús Carro
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 18 Mar 15
Posts: 275
Credit: 1,956,954
RAC: 22,420
Message 1724 - Posted: 15 Jul 2022, 8:03:47 UTC - in response to Message 1711.  

Hi boboviz,
We are working in the oposite direction. You are not simulating the effect of chemical compounds (drugs) to predict the behavior of the cardiac cells. We have experiments (e.g. from the University of of Szeged, and other from articles published in scientific journals) with which we can know how a cell responds to different drugs or situations. We need a model that is able to represent these results. There are some parameters that we can not measure in a cell, and we are trying to find the best value of them. So, we are simulating many combinations of these parameters to try to find the ones that better represent the experimental results.

In the process of fitting the model, we are not using all the data that we have. We have some data that later we will use to validate the result. The list of experiments is large, but, if you want I can post it.

Later, we will use the model to study pathologies and to try to predict what could happend, and yes in this case we will need experimental validation and we will need to colaborate with other groups.

Best,
Jesús.
Jesús Carro
Universidad San Jorge
@InSilicoHeart
ID: 1724 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[VENETO] boboviz

Send message
Joined: 9 Apr 15
Posts: 171
Credit: 1,483,869
RAC: 1,039
Message 1731 - Posted: 16 Jul 2022, 7:22:56 UTC - in response to Message 1724.  

Now it's very clear!
Thank you!!!
ID: 1731 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Yavanius
Avatar

Send message
Joined: 20 May 15
Posts: 50
Credit: 374,843
RAC: 428
Message 1733 - Posted: 16 Jul 2022, 16:04:12 UTC - in response to Message 1701.  

Good Morning.

So, I know it's early, but your current and beta models, have you noticed any differences (good and/or bad) over your original model?

Do you anticipate the volunteers needing to run the original model for comparison purposes?
ID: 1733 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Jesús Carro
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 18 Mar 15
Posts: 275
Credit: 1,956,954
RAC: 22,420
Message 1740 - Posted: 18 Jul 2022, 8:42:56 UTC - in response to Message 1733.  

So, I know it's early, but your current and beta models, have you noticed any differences (good and/or bad) over your original model?

We are testing the model with you because we have found good differences. We hope, at the end of the optimization to get good differences in others aspects, but at this point it is early and we will have to wait.

Do you anticipate the volunteers needing to run the original model for comparison purposes?

Yes! The previous model and other models previously published 🙂. We are preparing those simulations, we hope to start launching them shortly (1 or 2 weeks).
Jesús Carro
Universidad San Jorge
@InSilicoHeart
ID: 1740 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jaari

Send message
Joined: 9 Jul 22
Posts: 5
Credit: 798,097
RAC: 0
Message 1744 - Posted: 19 Jul 2022, 13:59:37 UTC

I noticed a variance in the names of the tasks, usually 1000, 2000, 3000, 4000, 5000, 6000, 10000 or no number. Is there anything specific between these tasks?
ID: 1744 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Jesús Carro
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 18 Mar 15
Posts: 275
Credit: 1,956,954
RAC: 22,420
Message 1746 - Posted: 20 Jul 2022, 6:29:27 UTC - in response to Message 1744.  

You are right, there is a bit of information in the names of the tasks. We usually call them with the first author of the protocol we are using or the name of the protocol. In some cases, we need to simulate the same protocol but varing a parameter. For example, in some cases we simulate the same protocol but stimulating the cell with different periods: 1000ms, 2000ms, 3000ms, etc. This permits to evaluate the model under different circunstances and to compare with the experimental results.

Best,
Jesús
Jesús Carro
Universidad San Jorge
@InSilicoHeart
ID: 1746 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile rilian
Avatar

Send message
Joined: 21 May 15
Posts: 16
Credit: 206,736
RAC: 3,846
Message 2363 - Posted: 10 Apr 2024, 20:24:00 UTC - in response to Message 1746.  
Last modified: 10 Apr 2024, 20:24:16 UTC

Hi Jesús

I was reading the https://denis.usj.es/denisathome/projects.php and I wonder about ongoing project progress, what is the current % done, how many tasks are expected, etc ? Could you please share some information..

Thank you very much!
--
I crunch for Ukraine

ID: 2363 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Jesús Carro
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 18 Mar 15
Posts: 275
Credit: 1,956,954
RAC: 22,420
Message 2374 - Posted: 16 Apr 2024, 7:46:02 UTC - in response to Message 2363.  

Hi Rilian,
The New human ventricular cell model (NHuVe) is now stopped because we have found some problems in the new model that requieres to go back to the definition and restart.

In the Human ventricular cell models optimization (HuVeMOp), we are currently optimizing one of the models indicated there (ORd). It is really difficult to estimate how much simulations we will need. Now we are using 4 different aproaches to find the best one with the ORd model. We will later continue with the other models. So, probably a lot of simulations remain...

Best,
Jesús
Jesús Carro
Universidad San Jorge
@InSilicoHeart
ID: 2374 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
magic_sam

Send message
Joined: 29 Apr 24
Posts: 2
Credit: 35,351
RAC: 3,213
Message 2385 - Posted: 29 Apr 2024, 18:55:31 UTC - in response to Message 2363.  

Hello there,

I second your suggestion @rilian :)

Some high-level information à la TN-Grid would do in my opinion:

https://gene.disi.unitn.it/test/gene_science.php

Best regards,

Samuel
ID: 2385 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile rilian
Avatar

Send message
Joined: 21 May 15
Posts: 16
Credit: 206,736
RAC: 3,846
Message 2388 - Posted: 30 Apr 2024, 21:09:18 UTC - in response to Message 2385.  
Last modified: 30 Apr 2024, 21:09:48 UTC

Hello there,

I second your suggestion @rilian :)

Some high-level information à la TN-Grid would do in my opinion:

https://gene.disi.unitn.it/test/gene_science.php

Best regards,

Samuel

Thanks!

Yeah, some projects display WUs numbers and % completion and it is always interesting to watch progress overtime and see the impact

Some other examples:
https://einsteinathome.org/server_status.php - see "search progress"
https://www.sidock.si/sidock/server_status.php - see "Research progress"
https://stats.distributed.net/projects.php?project_id=8
--
I crunch for Ukraine

ID: 2388 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Science : Previous Model vs New Model