Frenemies of TalkRational:
Nontheist Nexus |  Rants'n'Raves |  Secular Cafe |  Council of Ex-Muslims |  The Skeptical Zone |  rationalia |  Rational Skepticism |  Atheists Today | 
TalkRational  

FAQ Rules Staff List Calendar RSS
Go Back   TalkRational > Discussion > Life Science Discussions

Life Science Discussions Biology, Natural History, etc.

Reply
 
Thread Tools Display Modes
Old 02-21-2013, 09:12 PM   #2048421  /  #1
SteveF
Pleistocene person
 
SteveF's Avatar
 
Join Date: Mar 2008
Location: London, UK.
Posts: 27,175
SteveF
Default ENCODE

This is a pretty extraordinary paper:

http://gbe.oxfordjournals.org/conten....full.pdf+html

Some choice quotes:

Quote:
The ENCODE results were predicted by one of its authors to necessitate the rewriting of textbooks. We agree, many textbooks dealing with marketing, mass-media hype, and public relations may well have to be rewritten.
Quote:
Although we have failed to recruit 436 additional coauthors, we would like to express the hope that Jeremy Renner will accept the title role in the cinematic version of the ENCODE Incongruity
Quote:
So, what have we learned from the efforts of 442 researchers consuming 288 million dollars? According to Eric Lander, a Human Genome Project luminary, ENCODE is the “Google Maps of the human genome” (Durbin et al. 2010). We beg to differ, ENCODE is considerably worse than even Apple Maps.
SteveF is offline   Reply With Quote topbottom
Old 02-21-2013, 09:27 PM   #2048441  /  #2
Linus
opportunity of an erection
 
Join Date: Mar 2008
Posts: 4,887
Linus
Default

Graur also wrote the "reading the entrails of chickens" paper.
Linus is offline   Reply With Quote topbottom
Old 02-21-2013, 09:52 PM   #2048459  /  #3
MZBiologist
Member
 
Join Date: Jan 2010
Location: California
Posts: 72
MZBiologist
Default

An extraordinarily bad paper.

Their argument centers around their assertion that: "The ENCODE Consortium aims to convince its readers that almost every nucleotide in the human genome has a function and that these functions can be maintained indefinitely without selection."

This assertion immediately reminds me of the Republican "you didn't build it" attack on Obama.
MZBiologist is offline   Reply With Quote topbottom
Old 02-22-2013, 04:09 AM   #2048643  /  #4
Bilirubin
YOU HEINOUS TOOL
Bejeweled 2 Champion Adjunct
 
Bilirubin's Avatar
 
Join Date: Feb 2009
Location: SOME GODDAMNED METEOR FLYING AT LIGHT SPEED
Posts: 27,027
Bilirubin
Default

Quote:
Originally Posted by MZBiologist View Post
An extraordinarily bad paper.

Their argument centers around their assertion that: "The ENCODE Consortium aims to convince its readers that almost every nucleotide in the human genome has a function and that these functions can be maintained indefinitely without selection."

This assertion immediately reminds me of the Republican "you didn't build it" attack on Obama.
Would love it if you would ignore the stupid god bothering or whatever the atheist equivalent is here and elaborate your thinking here.
__________________
Quote:
Originally Posted by ksen View Post
Also, if you guys think I've been way out of line you can recall me.
Quote:
Originally Posted by the BEST elf View Post
i've always hated "well recall me then" as the only meaningful thing that can be done when a mod does stupid shit. it should not be an all or nothing proposition.
Bilirubin is offline   Reply With Quote topbottom
Old 02-22-2013, 04:12 AM   #2048644  /  #5
Flint
Senior Member
 
Join Date: May 2008
Posts: 1,181
Flint
Default

I enjoyed this paragraph especially:

Quote:
At this point, we must ask ourselves, what is the aim of ENCODE: Is it to identify every
possible functional element at the expense of increasing the number of elements that are
falsely identified as functional? Or is it to create a list of functional elements that is as
free of false positives as possible. If the former, then sensitivity should be favored over
selectivity; if the latter then selectivity should be favored over sensitivity. ENCODE
chose to bias its results by excessively favoring sensitivity over specificity. In fact, they
could have saved millions of dollars and many thousands of research hours by ignoring
selectivity altogether, and proclaiming a priori that 100% of the genome is functional.
Not one functional element would have been missed by using this procedure.
Flint is offline   Reply With Quote topbottom
Old 02-22-2013, 04:21 AM   #2048648  /  #6
sfs
Senior Member
 
sfs's Avatar
 
Join Date: Jan 2010
Location: Massachusetts
Posts: 420
sfs
Default

Mixed feelings here. I heartily applaud the excoriation of ENCODE's damaging hyping of their "functional" total. Also I like Dan Graur -- he helped me find some nice, reasonably priced chopsticks in Kyoto when I was shopping for presents -- and I find his over-the-top rhetoric amusing. On the other hand, his approach may make it easier for people to dismiss his arguments, and also leads him to spend too much time thwacking an easy target, rather than doing something more useful. ENCODE's purpose, after all, wasn't to determine what fraction of the genome was functional, but to provide lists of putative functional elements. What I would find useful is a balanced assessment of how good their various lists are. "Transcribed" as a category is clearly pretty much useless, but which of their other lists are swamped with false positives, and which are aren't? That's what determines whether ENCODE was a waste of money or not. Some of that information is in this paper, but it's not the focus.
sfs is offline   Reply With Quote topbottom
Old 02-22-2013, 05:23 AM   #2048674  /  #7
Flint
Senior Member
 
Join Date: May 2008
Posts: 1,181
Flint
Default

As far as I can tell, the focus of this paper is that the ENCODE approach was too inclusive for a variety of reasons. For example, their definition of "functional" was not the evolutionary notion, and even if it were, they assumed the consequent (if A implies B, therefore if we find B, we can assume A). I gathered that a great deal of what ENCODE regarded as "function" was derived that way. They tended to argue that if something clearly functional had certain properties, then everything they could find with those properties must be part of a "function".

So the overall impression I got was that (as I quoted) as a result, they narrowed things down very little. The difference between 9% and 80% is still too large to be useful.
Flint is offline   Reply With Quote topbottom
Old 02-22-2013, 12:59 PM   #2048740  /  #8
sfs
Senior Member
 
sfs's Avatar
 
Join Date: Jan 2010
Location: Massachusetts
Posts: 420
sfs
Default

Quote:
Originally Posted by Flint View Post
As far as I can tell, the focus of this paper is that the ENCODE approach was too inclusive for a variety of reasons. For example, their definition of "functional" was not the evolutionary notion, and even if it were, they assumed the consequent (if A implies B, therefore if we find B, we can assume A). I gathered that a great deal of what ENCODE regarded as "function" was derived that way. They tended to argue that if something clearly functional had certain properties, then everything they could find with those properties must be part of a "function".

So the overall impression I got was that (as I quoted) as a result, they narrowed things down very little. The difference between 9% and 80% is still too large to be useful.
The overall number won't be particularly useful anyway. The point of the project was to make catalogs of different classes of functional element, so that other studies that want to know, say, where all of the transcription factor binding sites near a certain gene are, just have to look it up. If one or two of their categories of "functional" are too broad and therefore useless, that doesn't affect the utility of their results much. If they all are too broad, then the whole thing was a waste of time and money.
sfs is offline   Reply With Quote topbottom
Old 02-22-2013, 07:08 PM   #2048965  /  #9
Steviepinhead
Senior Member
 
Steviepinhead's Avatar
 
Join Date: Mar 2008
Location: Seattle, WA
Posts: 29,993
Steviepinhead
Default

PZ weighs in here.
Steviepinhead is online now   Reply With Quote topbottom
Old 02-23-2013, 12:15 AM   #2049149  /  #10
teeth!
trollo trollini trollus est
Commissar
 
teeth!'s Avatar
 
Join Date: Mar 2008
Location: talkrationalo trollenda est
Posts: 46,823
teeth!
Default

Quote:
Originally Posted by Steviepinhead View Post
PZ weighs in here.
wow PZ has now taken to using the huffnpuff blockquotes when quoting scientific papers he disagrees with?

What a fucking fraud.
__________________
Quote:
Originally Posted by Schneibster View Post
Pro Tip™: I'M A CRAZY OLD BIGOT
Quote:
Originally Posted by Old Dan View Post
I tried [crack] once in September 1990 and it was nasty, it made me think "ew, this must be what bicycle tires taste like."
teeth! is offline   Reply With Quote topbottom
Old 02-23-2013, 12:19 AM   #2049150  /  #11
teeth!
trollo trollini trollus est
Commissar
 
teeth!'s Avatar
 
Join Date: Mar 2008
Location: talkrationalo trollenda est
Posts: 46,823
teeth!
Default

Also I read this response paper and I think some of the arguments made in the response are also shitty science. e.g. "we aren't seeing stabilizing selection, therefore it's not functional."

I can think of plenty examples of functional DNA that isn't undergoing stabilizing selection.
__________________
Quote:
Originally Posted by Schneibster View Post
Pro Tip™: I'M A CRAZY OLD BIGOT
Quote:
Originally Posted by Old Dan View Post
I tried [crack] once in September 1990 and it was nasty, it made me think "ew, this must be what bicycle tires taste like."
teeth! is offline   Reply With Quote topbottom
Old 02-24-2013, 01:05 PM   #2049651  /  #12
SteveF
Pleistocene person
 
SteveF's Avatar
 
Join Date: Mar 2008
Location: London, UK.
Posts: 27,175
SteveF
Default

[mod]moved some stuff[/mod]
SteveF is offline   Reply With Quote topbottom
Old 02-24-2013, 01:06 PM   #2049653  /  #13
SteveF
Pleistocene person
 
SteveF's Avatar
 
Join Date: Mar 2008
Location: London, UK.
Posts: 27,175
SteveF
Default

This is worth a read

http://blogs.scientificamerican.com/...itions-matter/
SteveF is offline   Reply With Quote topbottom
Old 02-24-2013, 04:34 PM   #2049694  /  #14
SteveF
Pleistocene person
 
SteveF's Avatar
 
Join Date: Mar 2008
Location: London, UK.
Posts: 27,175
SteveF
Default

also worth seeing ewan birney's tl

https://twitter.com/ewanbirney
SteveF is offline   Reply With Quote topbottom
Old 02-24-2013, 07:31 PM   #2049765  /  #15
Flint
Senior Member
 
Join Date: May 2008
Posts: 1,181
Flint
Default

Quote:
Originally Posted by SteveF View Post
This is worth a read

http://blogs.scientificamerican.com/...itions-matter/
I enjoyed Dan Graur's comment about the input from computer programmers. Bloatware surely has blocks of unreachable code, and these are a predictable product of the code maintenance process -- fixing bugs, adding patches, extending functionality, etc.

But more interesting is the category of useless code. This code is reached, can be quite extensive and complex, can produce plenty of results -- modify data structures, alter files, send messages, use up space and spend time. But over time, it comes to pass that nobody uses those data structures (or that part of them) anymore, or accesses those files, or receives the messages, etc. By almost any definition, this code is functional. And often enough, removing useless code breaks the program, because it had been inadvertently masking some unrelated mistake.

For example, rendering an uninitialized pointer harmless by having something harmless sit where the pointer ends up pointing. Or for example, placing traffic on the bus that masks what would otherwise be a race condition. Or for example, facing unreasonable deadlines a programmer might discover the value he wants is (for reasons he hasn't time to track down) always at offset N of a given file, so he wires that offset into the code.. Removing useless code can change that offset. And so on.

The point is, there's a very important distinction between useless and nonfunctional.
Flint is offline   Reply With Quote topbottom
Reply

  TalkRational > Discussion > Life Science Discussions

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT. The time now is 02:58 AM.


Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ©2008 - 2014, TalkRational.org