homeGeek CultureWebstoreeCards!Forums!Joy of Tech!AY2K!webcam

The Geek Culture Forums


Post New Topic  New Poll  Post A Reply
my profile | directory login | | search | faq | forum home
  next oldest topic   next newest topic
» The Geek Culture Forums   » Techno-Talking   » Science!   » Computational biology... I think...

 - UBBFriend: Email this page to someone!    
Author Topic: Computational biology... I think...
Stibbons
SuperBlabberMouth!
Member # 2515

Member Rated:
5
Icon 1 posted April 10, 2011 22:31      Profile for Stibbons   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
De-lurking to ask a question? How selfish of me...

Modelling physiological responses is fairly easy. Look up the equations, plug it into language of choice, run, and see if the answer is sensible. Yay, it is sensible.

But now I need to prove it is sensible by confirming the model responds in the exact same way as real life to the same physiological intervention. And I'm kind of stumped.

I was hoping with the combination of biology and computer geekery hanging around here, I may be able to find some clues as to how to go about this. Anyone got thoughts or experience with this kind of thing?

Posts: 1141 | From: UK | Registered: Dec 2003  |  IP: Logged
Ashitaka

SuperFan!
Member # 4924

Member Rated:
4
Icon 1 posted April 11, 2011 01:17      Profile for Ashitaka     Send New Private Message       Edit/Delete Post   Reply With Quote 
basically, you programmed your computer to say "owe" when you hit it, and you want to know how to prove this is a logical human responce.


Get a pad of paper and a pen, get out into the real world; and start expirementing!

--------------------
"If they're not gonna make a distinction between Muslims and violent extremists, then why should I take the time to distinguish between decent, fearful white people and racists?"

-Assif Mandvi

Posts: 3089 | From: Switzerland | Registered: Feb 2006  |  IP: Logged
Xanthine

Solid Nitrozanium SuperFan!
Member # 736

Member Rated:
5
Icon 1 posted April 11, 2011 05:30      Profile for Xanthine     Send New Private Message       Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Stibbons:
De-lurking to ask a question? How selfish of me...

Modelling physiological responses is fairly easy. Look up the equations, plug it into language of choice, run, and see if the answer is sensible. Yay, it is sensible.

But now I need to prove it is sensible by confirming the model responds in the exact same way as real life to the same physiological intervention. And I'm kind of stumped.

I was hoping with the combination of biology and computer geekery hanging around here, I may be able to find some clues as to how to go about this. Anyone got thoughts or experience with this kind of thing?

And down the rabbit hole a fine young geek slides...

Have you searched the literature to make sure no one has done this before?

If no, do that before you do anything else.

If yes...what's the system? Is it a simple metabolic pathway you can reconstitute in a test tube or do you need a living organism? If the latter, are we talking something nice and easy like E. coli or something insanely complicated and miserable like H. sapiens? If the latter, does it have to be something so complicated or would something simple like S. cerevisiae suffice?

--------------------
And it's one, two, three / On the wrong side of the lee / What were you meant for? / What were you meant for?
- The Decemberists

Posts: 7670 | From: the lab | Registered: Mar 2001  |  IP: Logged
Stibbons
SuperBlabberMouth!
Member # 2515

Member Rated:
5
Icon 1 posted April 12, 2011 18:48      Profile for Stibbons   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Xanthine:
Have you searched the literature to make sure no one has done this before?

Unfortunately not to this level. There's a lot of simulators out there, but none that have been tied down to reality precisely, instead preferring "it looks right, but if it acts crazy let us know.

quote:
If yes...what's the system?

It's something insanely complicated and miserable like H. sapiens, alas. We're poking patients undergoing pacing with drugs/changes to their pacemaker and seeing what happens to their physiology and ECG, essentially.
Posts: 1141 | From: UK | Registered: Dec 2003  |  IP: Logged
Xanthine

Solid Nitrozanium SuperFan!
Member # 736

Member Rated:
5
Icon 1 posted April 13, 2011 03:50      Profile for Xanthine     Send New Private Message       Edit/Delete Post   Reply With Quote 
Okay, so I like to work with molecules one molecule at a time so take this with a grain of salt...

The answer to your problem is statistics. Gather up a patient population, divide 'em into groups (don't forget a control!!), deliver treatments, and start observing. Once your study has run for as long as you've decided it should, sit down and do the math. I don't know enough about clinical research to even guess how many patients you'd need per group, but, when dealing with humans, more is better as you will be better able to control for other underlying factors, such as smoking or obesity or non-compliance with the meds or one of the million other things that are involved in heart troubles. However, more is also expensive. And you'll have institutional bureacracies and consent standards to deal with (are you beginning to see why I like molecules? [Wink] ). But, on the flip side, you can't take one patient and call them representative. People, as I'm sure you're all too aware of by now, are highly variable. You can't even take five and call them representative.

If the protocol you're studying is one that's already established and been in use for a while and you're just trying to develop a better predictive model, maybe you can get the in vivo results you need from case records. Accessing that information will involve red tape but it will be faster and cheaper than ramping up a whole new study from scratch.

Good luck. And find an actual clinician to talk to. [crazy]

--------------------
And it's one, two, three / On the wrong side of the lee / What were you meant for? / What were you meant for?
- The Decemberists

Posts: 7670 | From: the lab | Registered: Mar 2001  |  IP: Logged
GrumpySteen

Solid Nitrozanium SuperFan
Member # 170

Icon 1 posted April 13, 2011 07:56      Profile for GrumpySteen     Send New Private Message       Edit/Delete Post   Reply With Quote 
Xanthine wrote:
when dealing with humans, more is better as you will be better able to control for other underlying factors, such as smoking or obesity or non-compliance with the meds or one of the million other things that are involved in heart troubles. However, more is also expensive. And you'll have institutional bureacracies and consent standards to deal with

So, dump each drug into the water supply of a different major metropolitan area and monitor for changes in the number of people with pacemakers being hospitalizes or dying. Problems solved [evil]

--------------------
Worst. Celibate. Ever.

Posts: 6364 | From: Tennessee | Registered: Jan 2000  |  IP: Logged
Ashitaka

SuperFan!
Member # 4924

Member Rated:
4
Icon 1 posted April 13, 2011 08:25      Profile for Ashitaka     Send New Private Message       Edit/Delete Post   Reply With Quote 
quote:
Originally posted by GrumpySteen:
Xanthine wrote:
when dealing with humans, more is better as you will be better able to control for other underlying factors, such as smoking or obesity or non-compliance with the meds or one of the million other things that are involved in heart troubles. However, more is also expensive. And you'll have institutional bureacracies and consent standards to deal with

So, dump each drug into the water supply of a different major metropolitan area and monitor for changes in the number of people with pacemakers being hospitalizes or dying. Problems solved [evil]

lets not start the mandatory human testing BBQ again, my flamethrower is running low.

--------------------
"If they're not gonna make a distinction between Muslims and violent extremists, then why should I take the time to distinguish between decent, fearful white people and racists?"

-Assif Mandvi

Posts: 3089 | From: Switzerland | Registered: Feb 2006  |  IP: Logged
Xanthine

Solid Nitrozanium SuperFan!
Member # 736

Member Rated:
5
Icon 1 posted April 13, 2011 08:36      Profile for Xanthine     Send New Private Message       Edit/Delete Post   Reply With Quote 
quote:
Originally posted by GrumpySteen:

So, dump each drug into the water supply of a different major metropolitan area and monitor for changes in the number of people with pacemakers being hospitalizes or dying. Problems solved [evil]

Aside from that whole thing about informed consent, you also have issues with dosing.

--------------------
And it's one, two, three / On the wrong side of the lee / What were you meant for? / What were you meant for?
- The Decemberists

Posts: 7670 | From: the lab | Registered: Mar 2001  |  IP: Logged
Stibbons
SuperBlabberMouth!
Member # 2515

Member Rated:
5
Icon 1 posted April 13, 2011 08:44      Profile for Stibbons   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
Luckily my patient selection is fairly well controlled - they'll all be in ITU, so fairly easy to alter treatment with, and fairly easy to collect data. But yes, it will need multiple people per group, so statistics will become fun here. I'd best go hunt out an real life medical statistician (unfortunately the actual clinician said "student, sort it" [crazy] ).

Steen: I could if I would... [Wink]

Posts: 1141 | From: UK | Registered: Dec 2003  |  IP: Logged


All times are Eastern Time  
Post New Topic  New Poll  Post A Reply Close Topic    Move Topic    Delete Topic next oldest topic   next newest topic
 - Printer-friendly view of this topic
Hop To:

Contact Us | Geek Culture Home Page

© 2015 Geek Culture

Powered by Infopop Corporation
UBB.classicTM 6.4.0


homeGeek CultureWebstoreeCards!Forums!Joy of Tech!AY2K!webcam