7 replies to this topic

[Sanhar]

What are people's opinions regarding a "virtual human" model for testing and research purposes?

 

Some questions I have in particular:

 

1.  Is it feasible?  If it is, has research and development been done towards that end?

2.  How difficult would it be to create virtual versions of drugs to interact with this model?

3.  Could we accurately replicate the effects of genetic change on virtual cells?

4.  Could a project like SENS test their theories effectively on such a model?

5.  Would we be able to accelerate the effects on a virtual system to save time?

 

This is coming up as I am wondering given advances in computing if this would be possible as a means to both accelerate and reduce the costs of research while also upping accuracy and quality since mice etc. aren't exactly human and don't react quite the same way to testing.

 

I'm also reading about public opposition to animal testing and I wouldn't want that to derail our research efforts.  That being said killing droves of mice etc. isn't really ideal even if they are bred for that purpose so it's kind of hard to properly discount the opposition.

 

Thoughts please, thanks.

[Babol]

Sounds interesting at least. But how we can create it? 

[Mind]

This is already being worked on: http://www.longecity...ody-simulation/

 

There might be ways LongeCity could help out, but it seems the most feasible is just donating computing cycles (like folding@home).

[corb]

I read somewhere supercomputers powerful enough to simulate molecular biology will be feasibly cheap around 2080.

Feasibly cheap is still in the millions of dollars range mind you.

[Sanhar]

This is already being worked on: http://www.longecity...ody-simulation/

 

There might be ways LongeCity could help out, but it seems the most feasible is just donating computing cycles (like folding@home).

 

Speaking generally, the VPH is key to properly accelerating research.  Using mice etc. to do tests is a really undesirable method, not just because the mice suffer but because it isn't really very accurate.

 

"Organs on a chip" etc. as are coming out now are an improvement for sure as at least then it's actual human tissue but a VPH would be able to correlate massive amounts of human medical data to augment things AND the results could be quick, possibly near instantanous instead of taking months or years..  Some say we're going to need more advanced computing systems than we have now to really do this - and that may be true - but I am sure we can accomplish something even with current computer architecture and optical and quantum computers seem to be getting closer every day from what reports I am reading.

Edited by Sanhar, 04 April 2015 - 06:02 PM.

[Kalliste]

It will happen sooner or later. But it's possible that someone will have built a real AI by then 

 

Maybe our entire civilization is simply a brief simulation in someone elses computer. One that might end at any time.

Go read Permutation City by the way. It's obligatory reading for this kind of discussion.

http://www.goodreads...ermutation_City

[DeadMeat]

A simplified simulated mouse brain in a virtual mouse body from HBP. They still have a long way to go, to a realistic full mouse/human at molecular level, but the whiskers thing looks impressive.

 

And as for experimentation. Even if it is possible to create mouse/human level consciousness and pain with conventional computers. It can probably easily be turned off in a simulation, for what would otherwise be animal unfriendly experiments.

 

https://www.humanbra...tual-mouse-bo-2

 

As computing technology improves, their goal is to build the tools and the infrastructure that will allow researchers to perform virtual experiments on mice and other virtual organisms. This virtual neurorobotics platform is just one of the collaborative interfaces being developed by the HBP. A first version of the software will be released to collaborators in April.

 

The HBP scientists used biological data about the mouse brain collected by the Allen Brain Institute in Seattle and the Biomedical Informatics Research Network in San Diego. These data contain detailed information about the positions of the mouse brain's 75 million neurons and the connections between different regions of the brain. They integrated this information with complementary data on the shapes, sizes and connectivity of specific types of neurons collected by the Blue Brain Project in Geneva. A simplified version of the virtual mouse brain (just 200,000 neurons) was then mapped to different parts of the mouse body, including the mouse's spinal cord, whiskers, eyes and skin. For instance, touching the mouse's whiskers activated the corresponding parts of the mouse sensory cortex. And they expect the models to improve as more data comes in and gets incorporated.

[Sanhar]

Well, a virtual mouse is a good start.  Better than killing tons of real ones and faster data returns, too.