Print Topic - Archive

BTD Forums  /  (N=1).  /  *Beautiful Data
Posted by: Dr. D, Saturday, November 12, 2011, 8:01pm
http://generativemedicine.org/blogs/dadamolab/?p=23
Posted by: Lola, Saturday, November 12, 2011, 8:13pm; Reply: 1
humongous job Dr D!
bravo! :)
Posted by: RedLilac, Sunday, November 13, 2011, 3:47pm; Reply: 2
I had fun playing with it – not that I knew what I was actually reading about, but the geek in me liked it. ;D
Posted by: grey rabbit, Thursday, December 15, 2011, 1:34am; Reply: 3
very nice
Posted by: DoS, Wednesday, March 14, 2012, 1:49am; Reply: 4
I sent a PM asking about how some of those work. Just saying... since you probably get a few hundred a day.

Right now I am working on a paper for school concerning mapping. I am concentrating on the idea of holistic view, through algorithms as opposed to reductionist single sets of information within maps.
Posted by: ABJoe, Wednesday, March 14, 2012, 2:19am; Reply: 5
Quoted from DoS
I sent a PM asking about how some of those work. Just saying... since you probably get a few hundred a day.

Right now I am working on a paper for school concerning mapping. I am concentrating on the idea of holistic view, through algorithms as opposed to reductionist single sets of information within maps.

Can you glean any answers from the User's Guide where Dr. D. talks about some of the theory behind datapunk and Quod.?

http://www.datapunk.net/quodlibet/Quodlibet_User_Guide_1.pdf
Posted by: Spring, Wednesday, March 14, 2012, 2:44am; Reply: 6
Like everyone else, I'm glad I know this guy!!(smile)
Posted by: DoS, Wednesday, March 14, 2012, 2:05pm; Reply: 7
I will go through the user guide but I think my questions are a bit more than it is going to talk about.
Posted by: AKArtlover, Thursday, March 15, 2012, 6:56pm; Reply: 8
Quoted Text
The commonality between science and art is in trying to see profoundly – to develop strategies of seeing and showing. –Edward Tufte


<3 it! -- the quote and your work!
(sunny)
Posted by: Jane, Thursday, March 15, 2012, 7:19pm; Reply: 9
Not sure where to ask this but does Dr. D's post on FB today about autophagy and IBD (specifically Crohn's) mean that something like Trehalose is useful in treatment for those conditions?
Jane
Posted by: DoS, Thursday, March 15, 2012, 8:08pm; Reply: 10
The user guide was good, but my big question is still out there.
Posted by: ABJoe, Thursday, March 15, 2012, 8:23pm; Reply: 11
Quoted from Jane
Not sure where to ask this but does Dr. D's post on FB today about autophagy and IBD (specifically Crohn's) mean that something like Trehalose is useful in treatment for those conditions?
Jane

I think the multi-part post pin-pointed some of the cause as well as the cure.  For Secretors, much of the cause is harmful dietary lectins, so it is important to keep these out of the diet.  Another part of the cause is failure of autophagy induction (starting).  

Part of the cure, according to the article, is to stimulate autophagy.  Trehalose does this independent of the calorie restriction requirements of the MTOR gene.
Posted by: ABJoe, Thursday, March 15, 2012, 8:26pm; Reply: 12
Quoted from DoS
The user guide was good, but my big question is still out there.

Unfortunately, since the question was in a PM, no one else knows what it is, so can't begin to answer..
Posted by: AKArtlover, Thursday, March 15, 2012, 9:05pm; Reply: 13
http://metamodern.com/2010/09/15/trehalose-autophagy-and-brain-repair-sweet/
http://metamodern.com/2010/07/24/autophagy-why-you-should-eat-yourself/
http://www.ncbi.nlm.nih.gov/pubmed/20534972
Posted by: DoS, Thursday, March 15, 2012, 10:04pm; Reply: 14
I was essentially asking how his algorithm changes allow re-construction of networks as opposed to re-creation of them, by changing values. The idea being not to eliminate the first set of information but combine it to the second. Perhaps it is just be averaging, but I imagine it is more complicated involving cross-referencing etc.
Posted by: ABJoe, Thursday, March 15, 2012, 11:16pm; Reply: 15
Without access to the source code or knowledge of what modules Dr. D. has utilized from CPAN, I can't help with that...

I do know that we are copying the data from the Kegg maps into Quodlibet.  Once they are copied in, his software draws them, does the calculations he wants and links all of the information from the various databases to specific nodes.

I only go through this explanation to possibly answer that he is essentially recreating them in his software by copying the maps into Quod...
Posted by: Jane, Friday, March 16, 2012, 2:39pm; Reply: 16
Reading some of the studies cited above is confusing.  I'll just trust Dr. D. (sunny)
Posted by: Dr. D, Friday, March 16, 2012, 2:49pm; Reply: 17
Quoted Text
I was essentially asking how his algorithm changes allow re-construction of networks as opposed to re-creation of them, by changing values. The idea being not to eliminate the first set of information but combine it to the second. Perhaps it is just be averaging, but I imagine it is more complicated involving cross-referencing etc.


You're describing what are known as Bayesian Networks, where the  a posteri probablity is the starting point (plus any new data) for the a priori probability. This is a common computational algorithm in networks, but is not used the Datapunk platform.
Print page generated: Friday, April 25, 2014, 7:39am