More Recent Comments

Friday, September 07, 2007

Genomics Is Dead! Long Live Systems Biology!

 
When you're an old fuddy-duddy like me you've lived through several revolutions in biology. I still remember when recombinant DNA technology was going to change the world. Then it was developmental biology and evo-devo. Along the way were told with a straight face that sequencing the human genome would cure cancer and everything else.

After a while it all got very boring. We put up with the hype on the grounds that it was good spin framing for the general public. If it brought in lots of money then what's the harm? Well it turns out there was some harm done. We scientists are losing our credibility.

I've gone way beyond being bored by this kind of nonsense. Now I'm angry—especially when it seems that those who are ignorant of history are doomed to misrepresent it. Here's the opening paragraph of a press release on Systems Biology [Systems Biology poised to revolutionize the understanding of cell function and disease]. It summarizes the contents of a report to the European Science Foundation.
Systems Biology is transforming the way scientists think about biology and disease. This novel approach to research could prompt a shake up in medical science and it might ultimately allow clinicians to predict and treat complex diseases such as diabetes, heart failure, cancer, and metabolic syndrome for which there are currently no cures.
I wonder if they just reuse the reports from years past substituting "systems biology" for "genomics," or whatever the last cure for cancer was supposed to be? This kind of stupid motherhood hyperbole would be laughable if it wasn't for the fact that these people are deadly serious. That makes it pathetic.

Look what one of authors of the report has to say ...
Until recently, researchers tended to focus on identifying individual genes and proteins and pinpointing their role in the cell or the human body. But molecules almost never act alone. According to Lilia Alberghina from the University of Milano-Bicocca, Italy: “There is a growing awareness in medical science that biological entities are ‘systems’ – collections of interacting parts."
I suppose this depends on what you mean by "recently." If it's 40 years then maybe the statement might make some sense but even then it's a gross misrepresentation of the truth. Of course we isolated genes and proteins one-at-a-time but the goal was always to put them back together to make molecular machines. Does Lilia Alberghina really think that older scientists were completely unaware of the fact that biological entities are "systems"? I wonder if Alberghina is aware of metabolic pathways that were worked out half a century ago, or ribosomes, or DNA replication complexes, or muscle, or the complement system, or Drosophila embryogenesis, or any number of other systems that haven't just sprung into existence in the last few years.

Most scientist are already tired of these fads masquerading as revolution. I wonder how long it will be before the public and the politicians catch on?

78 comments :

A. Vargas said...

haha, Hye larry, systems biology is cool. of course some poeple get overnthusiastic and start tooting thir revolution horns.
I mena, what field does nto have these characters?
I fin francis crick and who was that other guy, hav made WAY more stupid declarations of reductionsit triumphalims.
And those guys DO move the moneys, ya know what I'm saying.

A. Vargas said...

"Sydney Brenner and Francis Crick held a seminar [in Cambridge, 1963] for the eight brightest students of the year to persuade them to become molecular biologists.
Brenner's pitch was this: "There are two great problems left to science. One is development and the other is consciousness. The reason that neither has been solved is that the people working on them are stupid. But now we have cracked the genetic code, we are going to finish them. Francis is going to take consciousness and I am going to take development. We will have them solved in ten years. Who wants to come along for the ride?"

hahahahahhahahaah

amazingly stupid deluded arrogant reductionist triumphalism; for all to see.

Anonymous said...

I agree, Sanders; systems biology IS cool. I have no doubt that its useful aspects will be assimilated into neo-Darwinism, enhancing our understanding of the materials that natural selection uses to shape adaptations.

A. Vargas said...

no, that will never happen. Becuase of the fagocitotic dogmatic appeal of selection as "ultimate" and cetral explanation.
Too many think already everything has been explained by selection. And they will continue to do so.
Plus, that fusion hasn't really happened for quite a long time now.

Joe Pickrell said...

systems biology IS cool. I have no doubt that its useful aspects will be assimilated into neo-Darwinism, enhancing our understanding of the materials that natural selection uses to shape adaptations.

absolutely. Uri Alon's brilliant book on systems biology ends thusly:
"Differences in biochemical details might be the result of historical accidents that are functionally neutral, or they might be governed by additional rules that have yet to be determined. One can always assume that certain differences are the result of historical accident, but such an explanation has no predictive power and tends to stifle the search for alternative hypotheses. It generally tends to be more productive if one starts with the working hypothesis that there are rules. One may end up attributing differences to historical accident, but in my opinion it is a mistake to start there."

A. Vargas said...

hahahha. yeah, some guys have really "mature" systemic thinking.
see why it has not been able to go anywhere? Other views are just much more easy

A. Vargas said...

"Alon takes a classic reductionist look at biological networks-- the whole can be understood by considering the parts"

So it's an anti-systems biology book.

Torbjörn Larsson said...

Oh, I think the public recognizes the overblown claims for research projects and that they are fads which misidentifies opportunistic work for grants. It is just easier for both parties to label them and pretend they are more coherent than they actually are.

So it's an anti-systems biology book.

Depends on your definition of reductionist.

The system analysis I read (dynamical systems) was reductionist in the sense of identifying local subsystems and their interconnection including feedback et cetera, where the subsystems in turn can be further broken down. Nested hierarchies in a sense. :-o Anyway, the whole is still understood by identifying its parts.

But I hear that biologists (Dennett, for one) have other views and descriptions. That could be great material for some future Sandwalk posts btw. Or perhaps Wilkins or p-ter already been there.

A. Vargas said...

When someone defends reductionism I lose all motivation to debate. it's like someone telling you that carrots are blue and you worrying badly becuasee you can't make them understand it's not. That is a complete waste of time with nothig in it for me.
I have no interest in proselitizing or defending systems theory., if I have mentioned it it's becuase with accustommed frivolity others have been pushed to label myself.
I do think systems theory is the best way of understanding biology. But that's just me. I think there is LOTS of work to be done, so I can understand the skepticism. Plus, everyone has his "extremists" that somehow manage to caricaturize and mislead the "begginers"

If you think already systems think systemes theory isa total stupidity, you obviously cannot expect me to "look forward" to explainign it to you. And truly, I don't.

Anonymous said...

Sanders, you're a piece of work. You give up when somebody "defends reductionism"? I think an approach that has produced all of physiology, Western medicine, biochemistry, molecular biology and genetics, and cell biology, not to mention big chunks of ecology, doesn't really need much defending. What kind of biology do you think is worthwhile?

Anonymous said...

What kind of biology do you think is worthwhile?

Meditating on the wholeness of being?

Strangely, he didn't seem to have problems with reducing all the variation in the antlers of different deer species to a simple relationship between antler length and body size. Damn that ultra-reductionist Steve Gould!

Joe Pickrell said...

The system analysis I read (dynamical systems) was reductionist in the sense of identifying local subsystems and their interconnection including feedback et cetera, where the subsystems in turn can be further broken down. Nested hierarchies in a sense. :-o Anyway, the whole is still understood by identifying its parts.

exactly. the next line in my review is, of course, "the parts, in this case, are not individual genes themselves, but the recurring types of regulatory interactions-- "motifs"-- that are seen in different types of network, biological or otherwise."

A. Vargas said...

Oh, os yhe parts are not really "parts"; this way we can still say the wole is still reducible to parts. yeah, whetever. Hahaha

Systems theory, specially in its biological roots, is not ony about the "control" cybernetic approach, wichi emphasizes hierarchies, etc. In the older biological tradition, emphasis is placed in emergent properties and epigenesis. Mayr and Gould emphasize stronly these emergent properties.
Reductionism negates any true emergence of new properties. It considers everyhting to as if an additive sum of the properties of the separate parts; only if this is true can we say that it's all about the parts.

"I think an approach that has produced all of physiology, Western medicine, biochemistry, molecular biology and genetics, and cell biology, not to mention big chunks of ecology, doesn't really need much defending"

I know what thsi is. "oh, if it's not reductionism, its not science!!" kind of phobia. Enmrgent properties magical, bad!! spooky spooky. Boo!

Anonymous said...

Sanders,

Anonymous said...

Sanders, what you fail to understand is that all of these things you keep bringing up as revolutionary alternatives to naive adaptationism - epigenesis, self-organization, fields, emergence - are not alternatives to adaptation via the accumulative action of natural selection. Rather, they serve to supercharge selection (in a sense, selection doesn't have to sweat the details) since selection can operate to change functional differences in developmental processes with supporting features being functionally facilitated by the structure of the system. Systems phenomena make selection all the more potent. Same with homeosis and other large epigenetic changes: more tools for selection to act on with other features to generate adaptive complexes.

The same is true of drift of neutral genes, biomolecules, and traits - as someone pointed out on another thread, they permit random walks across the adaptive landscape, exploring new potentials for adaptations. In the long run, neutralism is, like systems phenomena, an enabler of adaptation through natural selection.

Tupaia

Anonymous said...

Actually, all drift, not just drift in neutral entities, has that adaptation-facilitating property of random exploration of adaptive landscapes. It is also important to consider that the status of neutrality is often conditional on the organism and the environment, and so can become non-neutral.

Tupaia

Torbjörn Larsson said...

If you think already systems think systemes theory isa total stupidity

If I did, would I have taken that course?

Btw, pointing out a false choice in your analysis isn't the same as supporting the alternate choices.

Reductionism negates any true emergence of new properties.

Another false choice. (So by now you probably think I'm a super-reductionist. :-o) As in my description of the course, every level modeled had its own emergent properties.

For example, an amplifier may turn out to be of negative feedback type when you "open the box". But if you cut that feedback loop its properties change. The behavior emerges out of the system, and you can't understand it by considering any isolated part.

Another common observation is that quantum mechanics is an ill suited theory to describe macroscopic objects such as airplanes. Instead effective theories like hydrodynamics does a fair job on "emergent" collective properties. (But even so calculations are hard.)

But maybe I don't understand what you mean by "an additive sum" here. As I said, I would be happy to see an exposition of different definitions of reductionism and related concepts (and their reasons) in biology. As I understand it such discussions may have been much more active here than in other areas.

SPARC said...

From the Taxonomy of the Sciences over at The Daily Transcript:

"Systems Biologist:
System biologists are the newest branch on the life-sciences tree. No one is sure who, or what they are, only that some of them are ex-string physicists, others are bioinformatics guys, and the rest are biologists who have spent too much time playing Simcity or Civilization. Systems biologists collect data from all the other disciplines and then build their own fancy computer generated models. When they're done they head over to the pub and show off their fancy flow diagrams to the other biologists. Those systems guys try hard to impress the others, using words like "in silico", but no one pays any attention. If they're really desperate, systems guys will start using words ending with 'omics. Since systems biologists would love to kill, but not sure how to do it, they are mostly harmless.
(author: me)

Big Biologists:
Big Biologists were once little biologists. Some have even called themselves Systems Biologists. They saw the physicists with their particle colliders and rocketships. They saw their computer scientists with their huge mainframes. The little biologist became envious and started to ask "why can't we do big things?" They then started using Chips to analyze every gene. They crossed every mutant they had. "Think huge" First came genomics, then proteomics, then kinomics, ubiquitomics, phosphomics, p53omics. But in the end most of it was artifactomics.
I'm not sure what big biologists would kill, probably something really big, like the entire field of Biology.
(author: Pip)"

Anonymous said...

Haha. I agree with that description, which is enlightening about important defect of current systems theorists that have alienated them form mainstream biology.

Some guys have taken that avenue, of mathematization and computer models, as the way to "scientificize" the epistemological insights of systems theory. They DO sincerely think, as I do, that systems theory is a powerful thing, but it is true they are "physics-wannabes" and by reacing out immediatley for mathematical clarity, they have bypassed discussingany actual biology or natural history. So it's no wonder they cannot connect; and I must include myself among the non-impressed at the bar.
Systems theory must deal with natural history, testing its ideas against phylogenies, data from ecology, from development; contribute to understand the pathway taken in the evolution of specific, real lineages.
Our topics should draw form the same databases. we must deal with the topics that fascinate evolutionary biologists; speciation, adaptation, biogeography. From a systems perspective.

That's the challenge. Revolution? No, there will be no such thing. adaptationism and selection-worshipping will always be there to suck up the majorities,a seems to hav ben the situation for at least a century already.

I just think we can generate a small school of "systemic naturalists" that may add diversity to the intellectual ecosystem of volutionary theory.

Anonymous said...

Larsson:
"anyway, the whole is still understood by identifying its parts"

That to me is the basic reductionist statement that is off-limit stupid and uniteresting for me to discuss, but OK, I'll give you two lines of response. Haha.

You obviously need to pay attention to arrangement of the parts. A different arrangement of the same parts can give you a totally different whole. Identifying the parts clearly is insufficient.

A. Vargas said...

Consider that a diferent arrangement can be obtained by altering sewuence in which the parts fid each other in conforming that whole. If the properties of the parts make a simply additive contribution to the properties of the whole, this order of encounters, the "arrangement", could be whichever, and we would obtain the same result. Commutativity.

Anonymous said...

*shrug* Dude, I'm all about emergent properties: organisms, Integrative Physiology, like that.

Reductionism negates any true emergence of new properties.
See, that's where you're wrong. You can ask questions about some highly integrated system and emphasize emergent properties, or you can ask equally legitimate questions about causal mechanism and seek a "reductionist" explanation. And you can move up and down from level to level in the Organizational Hierarchy of Life as far as you want to in either direction. It's all OK. Let's drop the divisive false dichotomies that prevent our mutual understanding, brothers and sisters, and move forward under the blazing white Pluralist banner unfurled!

A. Vargas said...

On thing is to be analytical, to isolate and identify parts, modules. This is a greta thing in science.

Another entirely different thing is REDUCTIONISM; that the whole can be REDUCED to the parts; this is simply fase,but it continues to do great damage to biology specially through facile genetics: many biologists are more than happy to think there is a gene for anything.
In fact, many think that patts (genes) and not organism are the true actors of evolution, you know, the "tyranny of self replicators" envisioned by Dawkins...
Pus, I can assure you: with your attitude, so ecumenical and..lame, we will never squeeze out the juice form systems theory.

Anonymous said...

Sanders: "we will never squeeze out the juice form systems theory."

That's not quite the metaphor I have in mind. Rather, systems theory will be assimilated into the adaptationist colony like captured pupae. As were all of the rivals to adaptationism (Darwinism) before it.

Tupaia

A. Vargas said...

To make systems theory into a footnote to selection is not the way to encourage research into it.
Some silly selectionsits convince themselves they have assimilated everything. Not true. They hve just had to compromise when data is shoved into their faces. Acknowledging nod; then footnote. net result? Selectionists continue to be as igorant as ever about systems theory or development. "Synthesis", my ass.

Your long term, macroevoltuionary quasicosmic sense that all adaptaion and complexity is driven by natural selection may be very comforting to you bu to me it's just wooish and scientifically, completely useless. It seems to me that you are the kind of fool that needs to think he already understand everything by overrating a single factor (selection).

Anonymous said...

Saunders writes: Another entirely different thing is REDUCTIONISM; that the whole can be REDUCED to the parts; this is simply fase, but it continues to do great damage to biology specially through facile genetics: many biologists are more than happy to think there is a gene for anything.

The reductionism/holism schism promoted by some system biologists is simply stupid. It's "marketing" language used to sell the latest schemes and "superlabs" to non-scientists.

Saunders, systems biology *is* reductionistic as any other science. It is an attempt to understand the operation of biological systems as a phenomena of interactions. That's why I prefer the term, "integrative biology" to "systems biology". Just because the modelled systems are composed of more parts or are more "complicated" doesn't make it holistic or less reductionistic. Don't confuse "deterministic" with "reductionistic". The problem with many of the hoary overblown claims of genomics (DNA sequence analysis) was an assumption of determinism and oversimplification.

What gives integrative biology a shot at increasing our understanding is technological advancement. In somes cases there is now sufficient resolution of data to make the effort possible. At last, engineering and comp-sci approaches are becoming feasible. It's not a question of thinking holistically (try to convince an engineer with that notion!), it's a matter of bring additional scientific & demonstrated analytical resources to bear on biological questions.

It's turtles all the way down: Biology is chemistry is physics.

Anonymous said...

Clarification:

"It's turtles all the way down: Biology is chemistry is physics"

Meaning that as physical scientists, we try to connect the dots between levels. That doesn't mean that we can't find general rules and simplifications that allow us to address some questions at some level of detail while (sometimes) not worrying about *all* the interactions. (e.g. We can clone genes without solving the wave equation for a polymerase).

A. Vargas said...

Whos' talking about holism here? "holism" is a fitting word to describe the exactly symmetric absurd counterpart of "reductionism". We are not talking about butterfy effects here.

What's wrong with "determinism", please? How on earth could anyone be dumb enough to confuse it with reductionism?

"It's turtles all the way down: Biology is chemistry is physics"

Nope, and your "dot connecting" clarification is not enough. Biology does not violate chemistry, chemistry does not violate physicis; but none is encapsulated within the other. Each domain is unique and has its own rules. You ARE a reductionist.

Do you by any chance also think that organisms are "programmed" by their genes?

Anonymous said...

Sanders: Whos' talking about holism here? "holism" is a fitting word to describe the exactly symmetric absurd counterpart of "reductionism". We are not talking about butterfy effects here.

I can't parse that. Are you saying that holism is absurd, both terms are absurd, reductionism is absurd** or that critiquing the practice of integrative biology in terms of holistic/reductionistic dichotomy is absurd?

All I can say is that "systems thinking" is frequently described by some of its practioners as a more holistic approach. Those that do seem to think it gives the discipline a boost of some sort. Myself, I think it's just word-salad; simply a means of trying to differentiate one's research project in a crowd (As Larry would say, "trendy"). Like Larry, I've observed long enough to see how new trends in biological research get started (financially) and evolve. Integrative biology will certainly make useful contributions to our understanding of biology, whereas the metaphysical debate over how reductionistic a particular approach is probably will contribute little. Many scientists are practical people and go with whatever seems to work. For them, philosophical, navel-gazing debates about the nature of science is what happens *after* the work is done and after they've finished their third beer.

[...]
Nope, and your "dot connecting" clarification is not enough. Biology does not violate chemistry, chemistry does not violate physicis; but none is encapsulated within the other.

Not *entirely* encapsulated but there are problems in chemistry that are understood from the "standpoint" of physics, and there are biochemical problems understood in terms of physics and chemistry. There are some problems that can be effectively addressed while compartmentalized within a "domain" like chemistry or biology, but there are also cases of significant overlap.


Each domain is unique and has its own rules.

Sometimes, yes, and whose "media" depend on the interactions of the layers below...


You ARE a reductionist.

The term "straw man" comes to mind when I read that assertion.

** Please define "reductionist". Philosophic discussions without clear definitions are pointless.

Myself, I try to investigate problems with the appropriate tools for the job. I'm comfortable with the occasional black box. Of course, when a tool sometimes blows up in one's face, it's sometimes necessary to investigate the black boxes as well to make certain the underlying phenomena in separate instances really are translatable.


Do you by any chance also think that organisms are "programmed" by their genes?

Heh. I've studied metabolic regulation and am currently working with cancer biology. What do *you* think I think? Would anyone with that background suppose organisms are programmed exclusively by their genes?

Certainly some aspects can be "programmed" by introducing changes in some genes but the entire "program" doesn't reside exclusively in genes or their arrangement. Context matters. That was the problem with the hyperbolic claims of some of the previous genome sequencing enterprises: The assumption of genetic determinism and the belief that descriptions of genes are sufficient to determine the operations of an organism. At the 1st E.coli Genomic Meeting I heard one speaker say that a completed E.coli genome sequence would finally let us understand everything about the bacterium (He was a computer scientist and his comments went over like a lead balloon for the rest of the biologists). Lewtonin published a small book in 1991 that lashed out against that sort of thinking ("Biology as Ideology: The Dogma of DNA").

A. Vargas said...

I think the one who should come up with a defintion of reductionsim is yourself. I am just using the standard definition. Look up wikipedia or something if you don't know it. Jeez!

A. Vargas said...

You make the same stupid mistake Larry does. You have judged an epistemological posture without really understanding it nor looking into. Why? because 1) There has been no scientific revolution 2) someinvoke it vacuously to hype up research their and get money.
Therefore, you feel sure there is nothing to it. Just a fad.

A. Vargas said...

"Holism" ais frquntly extended beyond an anti-reductionsit meaning to argue an alleged great interconnetedness and underlying unity of everything (Spinoza, Hegel)
It is a term is used "holistic medicine", attached to the gaia hypothesis or in notions of It is also frequently invoke by people who think science should open its gates to religion, and stuff like that (for instance, Mr. Capra).
Make a google search for holism, a whole lot of woo will pop up. Not so if you search systems theory.

See why you are so easily fooled into thinking that systems theory is worthless? What would you know of guys like Von Foerster, Maturana?
Nuthin buddy. Just nuthin.

Larry Moran said...

Sanders said,

You make the same stupid mistake Larry does. You have judged an epistemological posture without really understanding it nor looking into. Why? because 1) There has been no scientific revolution 2) someinvoke it vacuously to hype up research their and get money.
Therefore, you feel sure there is nothing to it. Just a fad.


The so-called field of "systems biology" is a fad for two reasons.

1. The words "systems biology" are just a new term for something that we've been doing all along.

2. The claims of "systems biology" are just as silly as the exaggerated claims of all previous fads.

I'm not criticizing the idea of looking at the big picture. Instead, I'm criticizing the idea that this is something new and it's going to lead to a cure for cancer.

Anonymous said...

Jeez yourself, Sanders, you arrogant jackass. Take your own advice and look up "reductionism" on 'kipedia. The money quote: "There are several generally accepted types or forms of reduction in both science and philosophy"...See? Two people can use the same term correctly and still be talking about something quite different.

A. Vargas said...

"The words "systems biology" are just a new term for something that we've been doing all along"

No. As we can see right here, people do not have an accurate idea of what systems theory is (you yourself seem to think it is simply "looking at the big picture") when this perhaps is the most important philosophical topic of biology since aristotle: the exact relationship of parts and whole. Plus, may have a beatific view of reductionism and don't even understand what it actually menas; it is not analytical thinking; it is not "causality".

This is related to the depauperate philosophical knowledge of scientists who think they do not need to philosophize, just stare at data. That is, scientism.

A. Vargas said...

for instance, larry, do you think that lack of awareness about this topic has no relation to "evolutionary psychology" ?

A. Vargas said...

OK, a clarifictaion: a reductionist approach DOES work in some scientific situations. Which? it's obvious: when interactions are additive.
I am totally willing to admit that reductionism can be applied for instance, to explain the macrosopic shape of a salt crystal in function of its microscopic structure.
But not so the cell.

Actually, there are lots of situations in physics and chemistry where reductionsim is OK; but there are also many where it is not OK.

Do you guys think these distinctions are not worth making?

Jonathan Badger said...

Larry, honestly, "systems biology" is *not* something that "we've been doing all along", nor is it a fad. And I say this as a computational genomicist (who naturally competes with systems biologists for research funding and would love to criticize them if it was fair).

You have a point when you are complaining about how new techniques like systems biology (and yes, genomics too) are overhyped, but when you claim that systems biology is only a fad or nothing new you are just showing your own ignorance of the subject.

"Systems biology" isn't just platitudes about looking at organisms holistically, just like evo-devo isn't just banally observing that the developmental process had to evolve, I strongly recommend you read an introductory book on the subject such as Uri Alon's and then comment.

Anonymous said...

sanders writes (maybe to me): "Therefore, you feel sure there is nothing to it [integrative biology]. Just a fad.
and
"See why you are so easily fooled into thinking that systems theory is worthless?"

Funny. I'd previously written: "Integrative biology will certainly make useful contributions to our understanding of biology,..."

I'm against the hype, not the work. Grep some recent grant applications for the term "systems biology". In many cases it's exactly the same work the lab has always done.

Now I do disagree with Larry's assessment about systems biology being what we've done all along. It's an area we've always wanted to go but often lacked sufficient characterization of the basic biology and raw data necessary to go far. Attempted in the past, perhaps, but on a much smaller scale and with the realization of the limits. As such, I see it as an extension that many others (pointing to a couple UC Davis bacteriologists as early examples: John Ingraham and Allen "Jerry" Marr) have been pushing for a while behind the scenes. For example, Jerry Marr has long called for getting engineers interested in bacterial metabolism because of the different skill sets and approaches they would bring. The same for physicists.



*******************************

Also, Sanders writes: "OK, a clarifictaion: a reductionist approach DOES work in some scientific situations. Which? it's obvious: when interactions are additive.
I am totally willing to admit that reductionism can be applied for instance, to explain the macrosopic shape of a salt crystal in function of its microscopic structure.
But not so the cell.


So, when someone crunches expression correlations and develops an hypothesis of the contributions of a particular regulatory signal, how are they going to test whether this really is the case in the cell or organism and whether they've missed something? What are they going to diddle? Wouldn't they also explore how perturbations of underlying components affect the model? At some point the "rubber" of theory has to meet the "road" of real parts and real applications. It's an iterative process and it looks to my unexperienced eyes as a not untypical scientific approach (reductionistic or otherwise).

[...]
Do you guys think these distinctions are not worth making?

Honestly? Among the practicing scientists, most don't give a rats ass. What works, works. Those that use the wrong tool for the job are less likely to make good headway.

A. Vargas said...

I'm intrigued by this Uri Alon guy; in what precise sense does he defend reductionism? How much importance for biology is he willing to concede to reductionism?

I'm concerned because it is always typical that some one will try to compromise with new views by finding an argument to "assimilate" them but ALWAYS subdued to the previous view ( a typically fallacious exercise of lame scientific conservatism)

Is he actually stating that at the level of the entire organism, he can split it into modular networks such that the entire organism is just an additive sum of these modules?

I find that very difficult to believe; has he any evidence of this? Is he sure that if we rearrange those modules they woud render the same macroscopic result? You must convince me to buy this book!

Anonymous said...

Why not check out his publication list?

http://www.weizmann.ac.il/mcb/UriAlon/groupPapers.html

Tupaia

Anonymous said...

I'd forgotten the name but remembered one of the papers. Thanks for the reminder, p-ter and Tupaia.
U. Alon, M.G. Surette, N. Barkai, S. Leibler,
Robustness in Bacterial Chemotaxis
Nature 397,168-171 (1999).

Mel Simon's group published a related paper where they rearranged the kinetics equations into something that engineers would recognize as integrative feedback control.

Vol. 97, Issue 9, 4649-4653, April 25, 2000
Robust perfect adaptation in bacterial chemotaxis through integral feedback control
Tau-Mu Yi, Yun Huang, Melvin I. Simon, John Doyle

A. Vargas said...

He is modelling what we already know occurs; simple regulation, positive self-regulation. negative self regulation... then he is able to find his predicted motifs in experiments. OK, good enough!

What I do not understand is how does the discovery of these motifs mount up to a defense of reductionism; Nobody says that everything must interact in the organism: if everything did, maybe the expected patterns would not come up. But nobody can think that could be the case; if these motifs were not produced, we would never be able to experimentally observe a discrete effects from manipulating gene regulation experimentally.

Organisms are not homogenous continuities; they are conformed by parts and modules. Parts exist; they can be observed without problems. The fact that you can identify modules, even a whole set of them, does not imply reductionism.

Identifying these basic kinds of motifs is certainly an interesting and worthy scientific achievemnt but only on the error of an extended enthusiam could we make the mistake of thinking biology has thus been proven to be an ultimately reductionist science.

Does he actually do this in his book? I cannot really tell from his papers.

A. Vargas said...

Actually, we find cautious statements regarding what I consider relevant to my questions:

"These functions have been analysed using mathematical models and tested
with dynamic experiments in living cells. Still, there is much to be done: it is important to further experimentally
test the functions that each network motif can
perform. Such experiments could illuminate the
dynamics of the many systems in which each motif
appears. Furthermore, it is important to test whether motifs can help us to understand the densely connected networks of higher organisms"

A. Vargas said...

"At some point the "rubber" of theory has to meet the "road" of real parts and real applications. It's an iterative process and it looks to my unexperienced eyes as a not untypical scientific approach (reductionistic or otherwise)".

I think that what makes biology special lies much more in that "otherwise" than in the reducible, additive phenomena.

A. Vargas said...

"What works, works. Those that use the wrong tool for the job are less likely to make good headway"

Reductionism frequently doesn't work and is upheld as merely ideological reconfortation.

C'mon, UR. You are a data man, huh?

Give us a nice and explicit example of reductionsim succesfully applied in biology.

Larry Moran said...

Jonathan Badger wrote,

You have a point when you are complaining about how new techniques like systems biology (and yes, genomics too) are overhyped, but when you claim that systems biology is only a fad or nothing new you are just showing your own ignorance of the subject.

If you simply define "systems biology" as a "technique" then of course you're right. There are new techniques in science. Duh!

The old Department of Zoology here at the University of Toronto has renamed itself the Department Cell and Systems Biology. Is this sort of like calling itself the Department of Cell and SDS Polyacrylamide Gel Electrophoresis in your mind?

I don't think that's what you meant. Perhaps it was a slip of the fingers?

The outline of Uri Alon's book is here. There's nothing in the outline that refutes my earlier statement that systems biology is just the same old approach to understanding complex interactions that we've been doing for decades.

It's got a new name to make it sound sexy and there are some new techniques like microarrays and computer programs, but every field has new technology.

A. Vargas said...

systes theory is as much a fad as evo-devo is; by that I mena, it is something that must always be conssidered; no matter how many frivolous "political" arguments Larry may have.
Unforttunatey most biologists don't even get the point of the most elementary philosophy of systems theory, muh like many still don't get the point about the importance of development for the direction taken by evolution.

A. Vargas said...

all of this why you play dumb with the harm that the highy popular ideological reductionism DOES cause to biology

A. Vargas said...

some people make it look like a fad? well of course.
But true fads DO eventually disappear from the intellectual landscape (example: haeckelian biogenetic law as a principle for phylogenetic reconstruction)
But things systems theory and evo-devo just won't becuase they actually make an important point; and those who understand it cannot go back.

Larry Moran said...

Sanders says,

systes theory is as much a fad as evo-devo is

I agree.

Unforttunatey most biologists don't even get the point of the most elementary philosophy of systems theory

When I was a graduate student 35 years ago my first papers were about trying to reconstruct a molecular machine from its parts. We were well aware of the fact that the machine was more than the sum of its parts. My supervisor was one of the people who promoted the concept of molecular machines.

We weren't alone. You know the old cliché that those who are ignorant of history are history are doomed to repeat it? You are doomed.

A. Vargas said...

whatever. I think your attitude leads you to not being aware of the pervasiveness of bad reductionism in biology. You woudl worry abut that if you had any REAL systems awareness.You don't.

Jonathan Badger said...

Larry, studying network motifs as systems biologists do *is* completely new. Read the book. Of course, it may be deeper and more mathematical than you are used to, but that's part of the point. It isn't just a new "technique" any more than evolutionary biology is a "technique"-- it's a way of thinking.

I *know* that some traditional organismal biologists are trying to relabel their existing research programs as "systems biology", just as some geneticists once tried to relabel their work as "genomics", but that doesn't change the fact that genuine systems biology and genomics are more than relabeled zoology or genetics.

A. Vargas said...

For intance, Larry, do you agree with UR that reductionsim "works" and thus is just fine? You certianly seem to have abolutely no words of caution against reductionsim (which is typical of most biochemists, BTW)

YOu know what else "works"? adaptationism.

Reductionsim is as bad a lure as adaptationism.

Think about it.

A. Vargas said...

systes theory is as much a fad as evo-devo.
Only a foolish or ignorant person can think that evo-devo is a fad (it has actually alaways accompoanied evolution, except in the dark days hegemony of the "synthesis"), It is clear systems theory is "as much as a fad"
Both evo-devo and systems theory are here to stay. Like it or not.

Anonymous said...

Sanders writes: "Give us a nice and explicit example of reductionsim succesfully applied in biology."

Please define "reductionism" as *you* consider it. For example, you've read the papers from Alon and Simon on regulation of chemotaxis? Where do they fit on your scale? How does the tremendous amount of work that pulled apart and characterized the individual biochemical components to explain the observed phenomena rank on your scale?

***************************
Elsewhere Sanders writes: "Unforttunatey most biologists don't even get the point of the most elementary philosophy of systems theory,..."

"Philosophy of systems theory"? So "systems theory" isn't descriptive enough? Such an obsession with having the "right" philosophy...

On page 477 of is book, Science as Process, the author, David Hull attributes the following to Albert Einstein (from Einstein's book: Albert Einstein: Philosopher Scientist, 1949 ISBM: 0875481337):
"The reciprocal relationship of epistemology and science is of a noteworthy kind. They are dependent upon each other. Epistemology without contact with science becomes an empty scheme. Science without epistemology is - insofar as it is thinkable at all - primitive and muddled. However, no sooner has the epistemologist, who is seeking a clear system, fought his way through to such a system, than he is inclined to interpret the thought-content of science in the sense of his system and to reject whatever does not fit into his system. The scientist, however, cannot afford to carry this striving for epistemological systematic that far. He accepts gratefully the epistemological conceptual analysis; but the external conditions, which are set for him by the facts of experience, do not permit him to let himself be too much restricted in the construction of his conceptual world by the adherence to an epistemological system. He therefore must appear to the systematic epistemologist as a type of unscrupulous opportunist."

Hmm... "Unscrupulous opportunist". I *LIKE* that!

A. Vargas said...

*shrug* now you're just babbling and ultimatley show simple mistrust, moral accusations...My answer is: Whatever, dud. You're making a spectacle.

An example, dude!!! It's very simple. Don't send me to read yet more Alon. I will take you, or anyone, seriously and explain calmly to him why I think reductionsim doesnt't really work....despite the fact it..."works". jeje

BTW I ike that einstein quote very mcuh...but it really has nothing to do with systems theory, dude. You just can't use it any debate to your advantage, you know?

A. Vargas said...

Plus I think its obvious you are the only one here who seems to be pushing for some different notion of reductionism. That's OK!! let's hear it.

A. Vargas said...
This comment has been removed by the author.
A. Vargas said...
This comment has been removed by the author.
A. Vargas said...

When I say "philosophy", I wish to stress that I indeed do not need to be a mathematical genius, like Mr Alon for instance, to understand quite clearly why reductionsim is mostly inadequate for explaining biology.

Mr Alon's alleged ultimate "reductionism", notoriously absent in his papers (he does deal, after all, with little "systems" so it WOULD seem silly) must be in serious trouble if Mr Alon or any of his fans are unable to translate his thinking into the pretty clear terms of the philosophical discussion

Anonymous said...

James Watson:
http://www.02138mag.com/magazine/article/1488-2.html

"Only belatedly did Harvard try to enter the Genome Age by committing itself, as the 21st century began, to becoming strong in systems biology, a discipline so sprawling and unwieldy as to merit comparison to Enron in its limitless expansions before the collapse into nothingness."

I don't endorse the statement, but it is representative of the sentiments of many.

Tupaia

A. Vargas said...

As I said above, maybe guys like Alon may be partially responsible for this lack of any true revolution in the mentality of the majority of the scientific community. Guys like Alon dive head in to a pool of mathematics and computer modelation with phenomenological consequences that are not nearly dramatic enough for biologists to yet become some kind of "systems biomathematicians" (actually, I would not like that, haha).

I myself belive in the "one long argument" way: the biologists way. Big fat books like Darwin, Mayr and Gould, that explicitly adresses the natural history of life on earth and different sources of information.

In fact, immediate mathematization can "jump over" making some necessary philosophical distinctions and lost of progress constrained by our empirical knowledge of biology.

Torbjörn Larsson said...

Sanders:

Identifying the parts clearly is insufficient.

So you need observations and theoretical descriptions of the process. And that implies what?

Also, note that I described a course content to illustrate your false choice. Nothing else.

Commutativity.

Even reductionist theories like classical or quantum mechanics recognize commutativity. Rotations are commutative, for example.

Your description of reductionism as "additive sum" fails in an incredibly naive manner. I don't think you grasp reductionism (in any of the more common definitions) at all.

REDUCTIONISM; that the whole can be REDUCED to the parts;

Yes, but in many different ways, so called effective descriptions. And, maybe, some day at fundamental theory. There are no unique parts or descriptions, and by dualities not even for a fundamental theory.

There are more or less unique processes, at least at low energies where they have "frozen out".

Effective descriptions are emergent, btw.

facile genetics: many biologists are more than happy to think there is a gene for anything.

That isn't about reductionism, that is about model building (which should start with the simple) and lack of testing.

Now I'm disappointed. Is all the fuzz about a strawman?

Sven DiMilo:

You can ask questions about some highly integrated system and emphasize emergent properties, or you can ask equally legitimate questions about causal mechanism and seek a "reductionist" explanation. And you can move up and down from level to level ...

Hear, hear!

Torbjörn Larsson said...

Sanders:

I read the discussion on Alon.

FWIW, while I don't know what he does, there are several fields in physics where people "dive head in to a pool of mathematics and computer modelation" to either use toy models to parse out "phenomenological consequences" or to systematize phenomena and their data.

This is later useful in building theory or research methods. (Toy models are something physicists whips out to analyze new data, new phenomena, or new methods.)

A. Vargas said...

Larsson, I gave you the most clear example a dummy could want: You can build radically different machines with the same parts.

Please tell us exactly how in such case could you say that the parts explain the whole, in good ole raductiosnist fashion. It's a pretty simple questin, larsson. We shouldn't need to be mathematical geniuses to understad your answer, right?

"non-additive" and "emergence" are intuitive enough to help explain why this is possible. That you may quibble about the precise meaning of "additivity"? No surprise. So what then, "reductionism wins"? All I can say is: there is indeed such a thing as lack of philosophical agility.

"there are several fields in physics where people "dive head in to a pool of mathematics and computer modelation" to either use toy models to parse out "phenomenological consequences" or to systematize phenomena and their data"

yes, but what if the "phenomenological consequences" are not eloquentl?
It is easier to achieve phenomenological correspondence of math and models in physics and chemistry, where you mostly deal with much simpler systems.

However, it does not suprise me that the geniuses of physics may not care much about phenomenologiccal correspondece of mathematical models; After all, in modern physics the most briliant minds are engaged in string theory: an area notorious for its lack of phenomenological (experimental) grip.

This does not seem to work with biologists...and I'm glad

A. Vargas said...

make up you mind, Larsson. Your reductionism is :
a) "effective descriotion" (one of the stupidest arguments I have ever heard, BTW)
b) The "necessary simplest model" (as in gentic reductionism, that you DO defend. You're wrong)
c) "causal explanations", as you wer quick to agree with Sven about

Let me help you out:
d) all or any of the above so long as you can say "reductionsim prevails"

But I still recommend you strongly, that you try to at least understand the conventional philosophical notion of reductionism. Open a dictionary, encyclopedia or something, OK?

Anonymous said...

Sanders, to return to one of your earlier points, you have presented epigenesis as an alternative to adaptation through selection. But what phenomenon underlies much of the organizing principles of development? Selection, somatic selection - between individual cells, cell lineages, cellular connections, even molecules - for space, energy, reinforcement. (Cziko, Deacon, Edelman...) This somatic selection operates throughout life to shape organismic functioning, most famously in the brain and the immune system. But epigenesis and self-maintenance via selection is operating throughout the organismal system during both early development and the rest of the life cycle.

To turn to another "alternative," systems biology, it is not hard to find invocations of selection as shaping these interactive networks that systems biologists are so interested in.

Indeed, self-organization shares some important features with natural selection, including feedback that serves as an attractor towards a state of energetic/informational efficiency and stability.

A. Vargas said...

"But what phenomenon underlies much of the organizing principles of development? Selection, somatic selection"

Yeah, right; try to sell that one to the average developmental biologist or embryologist. good luck. Evne if any were willing to give it "lip service" selection is not a powerful tool for the study of development. Nor systems theory.

"To turn to another "alternative," systems biology, it is not hard to find invocations of selection as shaping these interactive networks that systems biologists are so interested in!

I have no doubt invocations of selection can be made for ANYTHING; don't you realize this is a PROBLEM, not a virtue? Notice that most of these eclectic selective invocations are mostly scientififcally vacuous, no matter how wonderful we may think their "explicative power" is.

Anonymous said...

I'm not saying that selection is the only game in town; I'm just pointing out its ubiquity at various levels of biological organization.

Tupaia

A. Vargas said...

yeah, well, water is always there, It's pretty ubiquous in biology.
It dos not mean that water explains biology, though, doesn't it.

A. Vargas said...

even if there would be no biology without it.

A. Vargas said...

"Even reductionist theories like classical or quantum mechanics recognize commutativity. Rotations are commutative, for example"

That's my point, you freakin "genius"

Anonymous said...

sanders writes: "*shrug* now you're just babbling and ultimatley show simple mistrust, moral accusations...My answer is: Whatever, dud. You're making a spectacle."

Well then, best of luck in your scientific career, sanders.

I'm not young enough to know everything - Oscar Wilde

A. Vargas said...

hey, dude, I'm still waiting for that example. Given you are such an "empirical", non-philosophical opportunistsic chap.
You're simply just no fun, UR.
Byebye

Torbjörn Larsson said...

Sorry for late response.

Sanders:

I will attempt an answer, in spite of your invectives. With some effort I can probably lower myself to your level.

You can build radically different machines with the same parts. Please tell us exactly how in such case could you say that the parts explain the whole, in good ole raductiosnist fashion.

I can't see that you have given an explicit example before.

In any case, we need to be more specific. Say that we have a resistance network connected to a voltage source. By changing the connections between the resistances, the current through the network will differ. (Example: parallel vs series connecting 1 Ohm resistances.)

The topology is different. Do you call that the same or a different machine? In any case, we can easily use Kirchoff's laws to model the network.

You are discussing philosophy, but since this is descriptive of how we see science works I don't see the relevance of your discussion. However, I was interested in a biologists description of reductionism. p-ter seems to have provided that. (Ie it is the usual. But how to define biological "parts" can be discussed.)

Your reductionism is

Again, this isn't "my" reductionism but a description of science.

- Effective field theories (effective descriptions) are our usual QFT's.
- "necessary simplest model" ???
- Physics is causal.

That's my point

Since I gave an example of how reductionism incorporates commutativity, "your" point is my point I believe. Is that really hard to understand?

A. Vargas said...

Larsson, if you cannot imagine how different machines could be made from the same parts, you are more stupid that I thought. I attribute this to the little imagination of your squarish, nearsighted "physicalist" kind; But don't expect much more, many times you are so flat-out stupid I truly wonder what are your motivations for "debating". It's not rational, you know.

If you truly do need an example, well here it goes: atomical composition, in percentage of each element, of a seagull and yourself is the same; C,H,O,N and "residues". There you go: Very different machines indeed; built with the same parts. If this does not help you, imagine lego houses or something.

Also: I'm saying commutativity is a property of reducible systems. So if you tell me there is commutativity in reductionist quantum mechanics all I can say is: great! What on earth did you think you were countering by pointing that out? There are PLENTY of cases where reductionism works in physicsand chemistry, I already said that.

Now if you tell me that reductionsim explains the difference between you and a seagull, I would say you're a dumbass. Why? because in this case, non-commutativity reflects how emergent properties and arrangement are crucial.

Your resistance network question is pretty easy: They are NOT the same machine. Kirchoffs law does not give you the arrangement. You will simply have to check that from a totally independent source of information before you say anything about the resistance of any such machine.

hen you say something as stupid as this:

"Again, this isn't "my" reductionism but a description of science"

Larrsson, you are so freakin cheap: now reductionism is "a descrition of science". Whta isin't reductionism isn't science. You have redefined science to be reductionsim. How can you lose?

You're wasting my time. Get lost.

A. Vargas said...

No matter how trivially simple the description, how basic the elements and their interactions: this is not equivalent to reductionism. Reductionism has a precise and sharp enough meaning; It explains the situation in which the properties of the parts trivially explain the properties of the whole (as in the salt cube example I borrowed from chemistry). In these cases we can safely say that the whole is all about the parts.

Obviously this is not the situation in the resistance network; arrangement is crucial as to the resistance of the whole. That example does NOT illustrate reductionism; the fact that we may describe the ssytem perfectly if we know the arrangment is an entirely different matter.

Are there any crunching sounds going on in your brain by now, Larsson?