Rugged fitness landscapes
Locked
 
|
In his PhD Thesis titled "Compositional Evolution", Richard A. Watson (now at Harvard University) argues that interdependencies between modules in the genotype of an evolutionary system are associated with the ruggedness of the fitness landscape, see http://www.oeb.harvard.edu/faculty/wakeley/richard/index.html (Table 1-1 on page 9, Figure 3-3 on page 83 of his thesis) * Weak interdependencies: smooth fitness landscape with a few optima * Modular interdependencies (between a few modules): a fitness landscape with some ruggedness * Arbitrary strong interdependencies: highly rugged landscape with many local optima Looks plausible, but is there hard evidence for it ? Is this true in general for evolutionary systems ? -J. |
|
|
Correction: he is no longer PostDoc at Harvard, has moved to Southampton http://www.ecs.soton.ac.uk/people/raw and his thesis is since February available at MIT Press as a book -> http://tinyurl.com/l4dc5 -J. |
|
|
In reply to this post by Jochen Fromm-3
Very interesting! It's too much to digest all at once but his summary
of simulation experiments on p240 of http://demo.cs.brandeis.edu/papers/watson_thesis_2002.pdf seems to show the same kind of step transition I found empirically in the G.tumida transition http://www.synapse9.com/GTRevis-2006fin.pdf, employing mechanisms not incompatible with those I proposed, fitting the logical necessities for the evidence of punctuated equilibria by providing a means for comprehensive change by a rapid process with a smooth start and end... Phil Henshaw ????.?? ? `?.???? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 680 Ft. Washington Ave NY NY 10040 tel: 212-795-4844 e-mail: pfh at synapse9.com explorations: www.synapse9.com > -----Original Message----- > From: friam-bounces at redfish.com > [mailto:friam-bounces at redfish.com] On Behalf Of Jochen Fromm > Sent: Monday, September 25, 2006 7:29 AM > To: 'The Friday Morning Applied Complexity Coffee Group' > Subject: [FRIAM] Rugged fitness landscapes > > > > In his PhD Thesis titled "Compositional Evolution", > Richard A. Watson (now at Harvard University) > argues that interdependencies between modules in > the genotype of an evolutionary system are associated > with the ruggedness of the fitness landscape, see > http://www.oeb.harvard.edu/faculty/wakeley/ric> hard/index.html > > (Table 1-1 on page 9, Figure 3-3 on page 83 > of his thesis) > > * Weak interdependencies: > smooth fitness landscape with a few optima > * Modular interdependencies (between a few modules): > a fitness landscape with some ruggedness > * Arbitrary strong interdependencies: > highly rugged landscape with many local optima > > Looks plausible, but is there hard evidence for it ? > Is this true in general for evolutionary systems ? > > -J. > > > ============================================================ > 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 > > |
|
|
Yes, his work contains interesting ideas. I especially like the appealing idea of correlating interdependencies in the genotype with the ruggedness of fitness landscapes, although it is probably to simple. Are interdependencies between modules in the genotypes a reason why small changes in the genotype could have large effects on reproduction rates ? It seems plausible, but hard to prove. The fitness function depends in general on the success of the phenotype (reproduction rates), and the relationship between genotype and phenotype is very complex and non-linear. The fitness of a phenotype is easy to determine, but hard to calculate from the genotype. This is similar to NP-complete problems: the quality of a solution is easy to verify, but the solution itself is hard to calculate. Therefore it is probably hard to say how rugged the fitness landscape is dependent on changes in the genotype, because the fitness is an unpredictable emergent property of the whole system, including the environment. One recent concept in this area seems to be "Epistasis" http://en.wikipedia.org/wiki/Epistasis Epistasis and Shapes of Fitness Landscapes http://arxiv.org/abs/q-bio.PE/0603034 -J. |
|
|
Well, it seems to open up to a wide variety of probably both feasible
and infeasible mechanisms. I think for small changes to have large effects there needs to be developmental process, i.e. a form of feedback of some sort. There are a *great* many possible means of establishing process feedback. That that mechanism, having results stimulate causes, has been ruled out of evolution theory for over a century seems to me to indicate a lack of imagination. Phil Henshaw ????.?? ? `?.???? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 680 Ft. Washington Ave NY NY 10040 tel: 212-795-4844 e-mail: pfh at synapse9.com explorations: www.synapse9.com > -----Original Message----- > From: friam-bounces at redfish.com > [mailto:friam-bounces at redfish.com] On Behalf Of Jochen Fromm > Sent: Tuesday, September 26, 2006 6:59 AM > To: 'The Friday Morning Applied Complexity Coffee Group' > Subject: Re: [FRIAM] Rugged fitness landscapes > > > > Yes, his work contains interesting ideas. I especially > like the appealing idea of correlating interdependencies > in the genotype with the ruggedness of fitness landscapes, > although it is probably to simple. Are interdependencies > between modules in the genotypes a reason why small changes > in the genotype could have large effects on reproduction > rates ? It seems plausible, but hard to prove. > > The fitness function depends in general on the > success of the phenotype (reproduction rates), and the > relationship between genotype and phenotype is very > complex and non-linear. The fitness of a phenotype > is easy to determine, but hard to calculate from the > genotype. This is similar to NP-complete problems: > the quality of a solution is easy to verify, but the > solution itself is hard to calculate. Therefore it is > probably hard to say how rugged the fitness landscape is > dependent on changes in the genotype, because the fitness > is an unpredictable emergent property of the whole system, > including the environment. > > One recent concept in this area seems to be "Epistasis" > http://en.wikipedia.org/wiki/Epistasis > > Epistasis and Shapes > of Fitness Landscapes http://arxiv.org/abs/q-bio.PE/0603034 > > -J. > > > ============================================================ > 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 > > |
|
|
Phil, indulge a layman for a moment: isn't auto-catalysis widely considered
to be the origination of life, and thus evolution? David dba | David Breecker Associates, Inc. www.BreeckerAssociates.com Abiquiu: 505-685-4891 Santa Fe: 505-690-2335 ----- Original Message ----- From: "Phil Henshaw" <[hidden email]> To: "'The Friday Morning Applied Complexity Coffee Group'" <friam at redfish.com> Sent: Tuesday, September 26, 2006 5:52 AM Subject: Re: [FRIAM] Rugged fitness landscapes > Well, it seems to open up to a wide variety of probably both feasible > and infeasible mechanisms. I think for small changes to have large > effects there needs to be developmental process, i.e. a form of feedback > of some sort. There are a *great* many possible means of establishing > process feedback. That that mechanism, having results stimulate causes, > has been ruled out of evolution theory for over a century seems to me to > indicate a lack of imagination. > > > Phil Henshaw ????.?? ? `?.???? > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > 680 Ft. Washington Ave > NY NY 10040 > tel: 212-795-4844 > e-mail: pfh at synapse9.com > explorations: www.synapse9.com > > >> -----Original Message----- >> From: friam-bounces at redfish.com >> [mailto:friam-bounces at redfish.com] On Behalf Of Jochen Fromm >> Sent: Tuesday, September 26, 2006 6:59 AM >> To: 'The Friday Morning Applied Complexity Coffee Group' >> Subject: Re: [FRIAM] Rugged fitness landscapes >> >> >> >> Yes, his work contains interesting ideas. I especially >> like the appealing idea of correlating interdependencies >> in the genotype with the ruggedness of fitness landscapes, >> although it is probably to simple. Are interdependencies >> between modules in the genotypes a reason why small changes >> in the genotype could have large effects on reproduction >> rates ? It seems plausible, but hard to prove. >> >> The fitness function depends in general on the >> success of the phenotype (reproduction rates), and the >> relationship between genotype and phenotype is very >> complex and non-linear. The fitness of a phenotype >> is easy to determine, but hard to calculate from the >> genotype. This is similar to NP-complete problems: >> the quality of a solution is easy to verify, but the >> solution itself is hard to calculate. Therefore it is >> probably hard to say how rugged the fitness landscape is >> dependent on changes in the genotype, because the fitness >> is an unpredictable emergent property of the whole system, >> including the environment. >> >> One recent concept in this area seems to be "Epistasis" >> http://en.wikipedia.org/wiki/Epistasis >> >> Epistasis and Shapes >> of Fitness Landscapes http://arxiv.org/abs/q-bio.PE/0603034 >> >> -J. >> >> >> ============================================================ >> 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 > |
Origin of Life (was: Rugged fitness landscapes)
|
|
David Breecker writes:
> Phil, indulge a layman for a moment: isn't auto-catalysis > widely considered to be the origination of life, and thus evolution? Harold Morowitz and Eric Smith have a very approachable working paper on Origin of Life: http://www.santafe.edu/research/publications/wpabstract/200608029 ABSTRACT: Life is universally understood to require a source of free energy and mechanisms with which to harness it. Remarkably, the converse may also be true: the continuous generation of sources of free energy by abiotic processes may have forced life into existence as a means to alleviate the buildup of free energy stresses. This assertion -- for which there is precedent in non-equilibrium statistical mechanics and growing empirical evidence from chemistry -- would imply that life had to emerge on the earth, that at least the early steps would occur in the same way on any similar planet, and that we should be able to predict many of these steps from first principles of chemistry and physics together with an accurate understanding of geochemical conditions on the early earth. A deterministic emergence of life would reflect an essential continuity between physics, chemistry, and biology. It would show that a part of the order we recognize as living is thermodynamic order inherent in the geosphere, and that some aspects of Darwinian selection are expressions of the likely simpler statistical mechanics of physical and chemical self-organization. |
Origin of Life (was: Rugged fitness landscapes)
|
|
Steve,
When I hear of gradients 'causing' self-organization .. as in "free energy by abiotic processes may have forced life into existence as a means to alleviate the buildup of free energy stresses".. I generally look for how the neutral pressures of gradients could dictate the intricate internal organization of the growth processes that develop from them. My own work began with the simplest sorts of conditions, the highly diffuse thermal gradients in the surface layers of fluids produced by sunlight and other heat sources. What you find there are amazingly intricate locally original and rapidly evolving systems of movement. One can easily accept that nothing would grow without gradients, but speaking of diffuse 'forces' causing intricately self-organized individual structures that spontaneously develop at irregular intervals is way more than a stretch. It's just dreaming. Gradients may produce instabilities in which complex systems are quite likely to develop, and life itself to appear to have come from one. It diverts all the good questions, though, to say it explains what happens inside the loops of evolving systems, or that the same cause of providing an opportunity for evolution must always produce the same effect. Phil > From: On Behalf Of Stephen Guerin > Sent: Tuesday, September 26, 2006 7:10 PM > To: 'The Friday Morning Applied Complexity Coffee Group' > Subject: Re: [FRIAM] Origin of Life (was: Rugged fitness landscapes) > > David Breecker writes: > > Phil, indulge a layman for a moment: isn't auto-catalysis > > widely considered to be the origination of life, and thus evolution? > > Harold Morowitz and Eric Smith have a very approachable > working paper on Origin of Life: > http://www.santafe.edu/research/publications/wpabstract/200608029 >ABSTRACT: Life is universally understood to require a source >of free energy and mechanisms with which to harness it. Remarkably, >the converse may also be true: the continuous generation of >sources of free energy by abiotic processes may have forced >life into existence as a means to alleviate the buildup of >free energy stresses. This assertion -- for which there is >precedent in non-equilibrium statistical mechanics and growing >empirical evidence from chemistry -- would imply that life >had to emerge on the earth, that at least the early steps >would occur in the same way on any similar planet, and that >we should be able to predict many of these steps from first >principles of chemistry and physics together with an accurate >understanding of geochemical conditions on the early earth. >A deterministic emergence of life would reflect an essential >continuity between physics, chemistry, and biology. It would >show that a part of the order we recognize as living is >thermodynamic order inherent in the geosphere, and that some >aspects of Darwinian selection are expressions of the likely >simpler statistical mechanics of physical and chemical >self-organization. ============================================================ 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 |
«
Return to Friam
|
1 view|%1 views
| Free forum by Nabble | Edit this page |