"Comparing genomes to computer operating systems in terms of the topology and evolution of their regulatory control networks"

[Russ Abbott]

See abstract below. The article is open source (http://www.pnas.org/content/107/20/9186.full.pdf+html) if anyone is interested.

The conclusions are

"The authors interpret these differences in terms of two design principles. The need for cost-effectiveness (or reusability) is central in programming, and robustness—that is, resistance to breakdown due to failure of a part—is the driving factor in biological systems. Evolution, they speculate, goes from top to bottom in software, but from bottom to top in biological systems."

I'm not sure I believe that either the comparisons or the conclusions are completely valid.  But it's an interesting comparison. Software evolves at all levels. But doesn't biology also? Aren't lower level functions perfected after being incorporated into higher level entities?  It seems to me that biology is just messier and less well designed.  No one refactors biological systems. But it seems like the redundancy produces more robustness.


-- Russ

---------- Forwarded message ----------
From: Science Editors' Choice <[hidden email]>
Date: Thu, Jun 3, 2010 at 12:16 PM
Subject: Science CiteTrack: Editors' Choice: Highlights of the recent literature
To: [hidden email]Systems Biology:

E. coli (left) and Linux (right) networks

CREDIT: KOON-KIU YAN AND NITIN BHARDWAJ

Yan et al. have compared the transcriptional control network in the bacterium Escherichia coli to the network depiction (known as the call graph) of the Linux kernel, which is the central component of a highly popular operating system. Both systems feature (i) master regulators (yellow in the graphic), which send directions to targets; (ii) middle managers (red), which both send and receive orders; and (iii) workhorses (green), which are controlled but do not control others. For the bacterium, there are lots of workhorses but relatively few regulators at the other levels. The Linux call graph is top-heavy or more populated at the master regulator and middle-manager levels. In other words, a workhorse in the transcriptional network usually has only a few supervisors, but in Linux, a workhorse answers to a large number of regulators. The authors also contrasted evolution in the two systems by looking at the functions that persist in 24 versions of the Linux source code relative to genes that persist in 200 phylogenetically distinct bacteria. For E. coli, the workhorses showed the greatest persistence, whereas for Linux, there was persistence at all three levels, but mostly in the master regulators and middle managers. The authors interpret these differences in terms of two design principles. The need for cost-effectiveness (or reusability) is central in programming, and robustness—that is, resistance to breakdown due to failure of a part—is the driving factor in biological systems. Evolution, they speculate, goes from top to bottom in software, but from bottom to top in biological systems.

Proc. Natl. Acad. Sci. U.S.A. 107, 9186 (2010).

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

[Grant Holland]

You might enjoy this related quote from Jacques Monod - Nobel laureate and considered by many as the "father of molecular biology":

"We call these [mutation] events accidental; we say that they are random occurrences. And since they constitute the only possible source of modifications in the genetic text, itself the sole repository of the organism’s hereditary structure, it necessarily follows that chance alone is at the source of every innovation, of all creation in the biosphere." [Monod 1972]

I submit that the software engineering is far from a "chance alone" endeavor.

Grant

Russ Abbott wrote:

See abstract below. The article is open source (http://www.pnas.org/content/107/20/9186.full.pdf+html) if anyone is interested.

The conclusions are

"The authors interpret these differences in terms of two design principles. The need for cost-effectiveness (or reusability) is central in programming, and robustness—that is, resistance to breakdown due to failure of a part—is the driving factor in biological systems. Evolution, they speculate, goes from top to bottom in software, but from bottom to top in biological systems."

I'm not sure I believe that either the comparisons or the conclusions are completely valid.  But it's an interesting comparison. Software evolves at all levels. But doesn't biology also? Aren't lower level functions perfected after being incorporated into higher level entities?  It seems to me that biology is just messier and less well designed.  No one refactors biological systems. But it seems like the redundancy produces more robustness.


-- Russ

---------- Forwarded message ----------
From: Science Editors' Choice <[hidden email]>
Date: Thu, Jun 3, 2010 at 12:16 PM
Subject: Science CiteTrack: Editors' Choice: Highlights of the recent literature
To: [hidden email]Systems Biology:

E. coli (left) and Linux (right) networks

CREDIT: KOON-KIU YAN AND NITIN BHARDWAJ

Yan et al. have compared the transcriptional control network in the bacterium Escherichia coli to the network depiction (known as the call graph) of the Linux kernel, which is the central component of a highly popular operating system. Both systems feature (i) master regulators (yellow in the graphic), which send directions to targets; (ii) middle managers (red), which both send and receive orders; and (iii) workhorses (green), which are controlled but do not control others. For the bacterium, there are lots of workhorses but relatively few regulators at the other levels. The Linux call graph is top-heavy or more populated at the master regulator and middle-manager levels. In other words, a workhorse in the transcriptional network usually has only a few supervisors, but in Linux, a workhorse answers to a large number of regulators. The authors also contrasted evolution in the two systems by looking at the functions that persist in 24 versions of the Linux source code relative to genes that persist in 200 phylogenetically distinct bacteria. For E. coli, the workhorses showed the greatest persistence, whereas for Linux, there was persistence at all three levels, but mostly in the master regulators and middle managers. The authors interpret these differences in terms of two design principles. The need for cost-effectiveness (or reusability) is central in programming, and robustness—that is, resistance to breakdown due to failure of a part—is the driving factor in biological systems. Evolution, they speculate, goes from top to bottom in software, but from bottom to top in biological systems.

Proc. Natl. Acad. Sci. U.S.A. 107, 9186 (2010).

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

[Douglas Roberts-2]

I've seen software engineering initiatives where the decision process was positively Brownian.  Participants darting first in one direction, and then just as unpredictably vectoring off in a completely different random direction.

Of course I've also enjoyed watching the "Keystone Kops" software engineering methodology.  Hugely entertaining, but not much more effective than the purely random method at the end of the day.

It would be interesting to learn how many software initiatives world-wide fail in their first few years.

--Doug

On Thu, Jun 3, 2010 at 5:15 PM, Grant Holland <[hidden email]> wrote:

I submit that the software engineering is far from a "chance alone" endeavor.

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org