Excessive reductionism in Biology?

[ryokan]

One of keys in the development of the modern Biology, exemplified by the advances in Genetics was due to a reductionist view.
Nevertheless it is possible that an excess of reductionism have side effects. So, face to Biochemistry, less work was done in Biophysics (reflected in databases as PubMed), excepting some fields such as the biological effects of ionizing radiations. Likewise, besides a lot of work in Genetics/Genomics and, more recently, Proteomics, less research is being conducted on the physical properties of the cell.
There are interesting perspectives, such as the complex systems theories, centered on the so called emergent properties. But I believe that an excess of reductionism can blind many interesting fields of basic and applied research.

 

[loseyourname]

I do think there has been a bit too much focus on molecular biology. It seems almost every bio-student I know is going for molecular-bio. I know the field is really hot in the wake of the Human Genome Project, but it's going to eventually reach saturation in terms of employed researchers and other disciplines are being neglected. Take entomology, for instance. I doubt many people here realize that many of the best paid biologists in the US are entomologists, simply because they are so rare and are very important to agriculture.

 

[Moonbear]

I agree as well. I use some molecular approaches, but keep my feet solidly grounded in systems. I run into too many molecular biologists who will stare at me blankly when I ask something like, "but what does it do in the animal?" The old adage "can't see the forest for the trees" comes to mind. I think both approaches are good, but at some point you have to meet in the middle to make either useful.

 

[iansmith]

Carl Woese started a similar discussion with a paper. I try to start the same topic. I still trying to work some notes out the paper.

https://www.physicsforums.com/showthread.php?t=34411

 

[Moonbear]

Is that what that paper was about? I recall being tired the night I read it, but also remember it not making a lot of sense the way it was written. The two may have been related (my tiredness and inability to make sense of the article).

 

[iansmith]

The paper is in part about the redutionism of molecular biology and it effect biology, and it is also an autobiography. Sometimes it get confusing and he review some of previous paper which I have read and I skip ahead for these part.

 

[ryokan]

I do think there has been a bit too much focus on molecular biology. It seems almost every bio-student I know is going for molecular-bio. I know the field is really hot in the wake of the Human Genome Project.

Besides a basic scientific interest on Molecular Biology, there are powerful economic factors. The development of new therapies semms to have a strict reductionist view. The old sentence "one gene one enzyme" is, in practice, extended to "one gene, one disease" in the search for therapies. I remember the hopes arised from the finding of both monoclonal antibodies (magic bullets) and gene therapy. Now, they are applied to only a few diseases. Probably, whereas Pharmaceutical Industry remains reductionist, Biology will be in practice only Mol Biol.

 

[Adrian]

Besides a basic scientific interest on Molecular Biology, there are powerful economic factors. The development of new therapies semms to have a strict reductionist view. The old sentence "one gene one enzyme" is, in practice, extended to "one gene, one disease" in the search for therapies. I remember the hopes arised from the finding of both monoclonal antibodies (magic bullets) and gene therapy. Now, they are applied to only a few diseases. Probably, whereas Pharmaceutical Industry remains reductionist, Biology will be in practice only Mol Biol.

'When all you have is a hammer, everything looks like a nail.' I mean, we are at a point were we can use use genetics and molecular biology for diagnosis or drug development in diseases associated with only one or a very limited number of genes. So what we do is get really excited about the new possibilities in meidcine, and then look for diseases that are caused that way (by one or very few genetic defects) and try to use our molecular biological methods to cure, treat, or at least diagnose them and understand them better.

Whereas in other cases - take cancer, for instance - we tried it that way and saw that it doesn't work this simply. So what we do is do soem more research, put together more pieces of the puzzle, gain an ever better understanding of the thing - and at some point in the future, we will have figured it out enough to take our knowledge and our improved methods and develop a cure for it.

We're still *very* far away from knowing enough about the molecular interactions at the base of life and having methods that are precise enough in order to explain complex systemic phenomena in a synthetic manner, drawing upon the molecular parts. But that doesn't mean this approach is not viable; on the contrary, the fact that we still have so much work to do is an argument for focussing quite a bit of efforts on molecular biology.

Having said that, it is certainly true that because of the genetics hype, a large number of students just go into the field because it's "hot", and there surely are more aspects of biology that are very interesting as well.

Or am I mistaken?

 

[ryokan]

But that doesn't mean this approach is not viable; on the contrary, the fact that we still have so much work to do is an argument for focussing quite a bit of efforts on molecular biology.

Having said that, it is certainly true that because of the genetics hype, a large number of students just go into the field because it's "hot", and there surely are more aspects of biology that are very interesting as well.

Or am I mistaken?

No. You are right. The advances in basic Mol Biol are impressive. And their applications to diagnosis in Human Genetics, among others, are astonishing. New methodologic advances, microarrays, use of iRNA, and so on, promise a future very wealthy in both basic and applied findings. I think that is obvious.
As you write, "we still have so much work to do". I agree.

But I think that there is an excess of reductionism. For example:
a) In the importance done to the gene expression in "structural" terms: sequencing and differential expression (tissue or pathology - related) face to a more scarce interest on non-codifying DNA and the environmental influences and dynamic variations. What about junk DNA?
b) A relative scant research on Biophysics. The influence of temperature on cells is only marginally studied (heat shock proteins). As a little example, I think it is very curious the so called reaction norms of phenotypic traits, as that seen in Drosophila.
c) The relative paralysis of research in other traditional biological fields

I believe that now everything is being "reduced" to DNA.

 

[Adrian]

a) In the importance done to the gene expression in "structural" terms: sequencing and differential expression (tissue or pathology - related) face to a more scarce interest on non-codifying DNA and the environmental influences and dynamic variations. What about junk DNA?

Great point, and I agree that this is an area that deserves a lot of attention in the future. So far, we've been quite busy sequencing a few genomes in the first place, identifying genes and then determining what portions of the DNA are 'junk DNA' at all. So now to the tricky part, so to speak. Personally, I find it extremely intriguing how much we can (perhaps) learn from junk DNA about the workings of evolution. Also, when it comes to applied science, we could improve the effectiveness of our methodology greatly if we figured out just how exactly (particular) genes are regulated in the human cell so that gene therapy requiring regulation would no longer depend on large genomic transgenes, to give just one example. This also illustrates IMO the major reason why many of the bright promises of genomics haven't (yet) been put to successful use in applications: In relation to the mechanism of nature, our methods of intervention are still rather primitive (adenoviruses in gene therapy or using gold particles to genetically modify eukaryotic cells are other examples for methods that are not nearly as refined and accurate as they would have to be in order for us to be able to actually apply the theoretical knowledge we have).
So, in short, I absolutely agree.

b) A relative scant research on Biophysics. The influence of temperature on cells is only marginally studied (heat shock proteins). As a little example, I think it is very curious the so called reaction norms of phenotypic traits, as that seen in Drosophila.
c) The relative paralysis of research in other traditional biological fields

This probably does have to do a lot with the rush into genomics and proteomics, but I have a little problem with the term "reductionism" as the culprit's name. It's been the claim of a great number of people with an anti-science mind set that biology's attempts to explain complex phenomena - be it the origin of species or the nature of mind and consciousness - are undue reductionism, and that emergent properties are not a viable explanation. That's why I'm uncomfortable with the term. But if you say that the DNA hype has led to the unfortunate neglect of other biological fields (which, if I understand correctly, is indeed what you're saying), I agree fully.

Alas, maybe I have no room to talk, since I am one of those students who are beginning their studies with a focus on molecular bio these days myself...

 

[ryokan]

I have a little problem with the term "reductionism" as the culprit's name. It's been the claim of a great number of people with an anti-science mind set that biology's attempts to explain complex phenomena - be it the origin of species or the nature of mind and consciousness - are undue reductionism, and that emergent properties are not a viable explanation. That's why I'm uncomfortable with the term. But if you say that the DNA hype has led to the unfortunate neglect of other biological fields (which, if I understand correctly, is indeed what you're saying), I agree fully.

Yes. "Reductionism" is also used in an anti-science sense. Of course, that isn't the case here. I think we agree with the use of this term in Biology in the sense that we have wrote in our posts.
Do you suggest another, more useful, term?

 

[ryokan]

Carl Woese started a similar discussion with a paper. I try to start the same topic. I still trying to work some notes out the paper.

https://www.physicsforums.com/showthread.php?t=34411

I have read the article of Woese in PNAS. It is really interesting the attention to horizontal gene transfer (HGT) in evolution (in origin, it would be non-darwinian) and the suggestion of Woese that "cellular evolution must begin in a collective mode". I understand that, following this theory, cells and darwinian evolution would arise when HGT is, in practice, impossible. Before, we would have only "supramolecular aggregates". If so, a key factor to the "Darwinian Threshold" could be the evolution of membrane systems.

 

[ryokan]

DNA languages

The actual reductionist view in Mol Biol can partially result from the consideration of DNA as the support of genetic code. Of course that is true. But the great advances that this discovery made possible have difficulted the wide view of the use of other languages by DNA.
Languages of direct communication both with proteins (with a regulator or structural role) and even DNA ("communication" among regions with transposons). From junk DNA we have already discussed in this forum.

 

[ryokan]

Comparative Biochemistry and Physiology

One possible consequence of the reductionist viewpoint is the scarcity of work devoted to the study of comparative biochemistry and physiology.
Nevertheless, a lot of knowledge could derive from such studies. For example, from differences in resistance to diseases among different animals, or studying why some plants make medicinal drugs.

 

[Moonbear]

One possible consequence of the reductionist viewpoint is the scarcity of work devoted to the study of comparative biochemistry and physiology.

Yes, this is true. The potential consequence is having too few scientists trained in these other areas, which will become extremely important as genes are identified and modified and we need people to see if they have any effects outside that targetted. Then again, this is common in the sciences. A particular "hot topic" gets all the interest, money, new buildings, influx of students, and eventually it is replaced by a different hot topic. In the short term, you can do well in one of those hot fields, but then things level out because you have too many qualified people in the field and not enough jobs for them. On the other hand, the fields that are not so "hot" get fewer students, who wind up in much greater demand in the job market. Hot fields can shift in just a few years time, so it's really hard to predict anything when you start out. Just pick what you enjoy doing.

 

[ryokan]

What moves hot topics?

Yes. The change in hot topics is clear.
But, what moves hot topics? Only scientific interest? I think that Molecular Biology is directed in great measure by Pharmaceutical Industry.

 

[ryokan]

quorum sensing

A question on reductionism would be: is it ever possible?
Such would be the case with quorum sensing in bacteria. The behavior of a group of bacteria, both in vivo (biofilms) or in vitro (colonies on agar) seems to be very most complicated than a sum of individual actions.
For this example an other similar cases, my question would be:
Are there fundamental limits to a reductionist viewpoint in Biology ?

 

[Moonbear]

Are there fundamental limits to a reductionist viewpoint in Biology ?

I think the answer is a resounding YES. Not everything happens at the level of the gene. There are quite complex interactions at the protein, cell, and organismal levels. Sequencing the entire genome tells us nothing about how the organism functions. We've moved from just genomics to proteomics, and I think the next step must be functionomics (a.k.a., physiology - but it seems to be cutting edge nowadays, you have to add "omics" to the end of some word).

 

[selfAdjoint]

But it's pretty much that protein structure arises given a genome structure, and there's a lot of research going on in this, both pure protein (folding, etc), generation (RNA, etc.) and genome (timing, promoters and inhibitors, transposon controls, etc. etc.). This is all good new knowledge, which would be missed if we just sat back and said "emergent, emergent"

 

[ryokan]

But it's pretty much that protein structure arises given a genome structure, and there's a lot of research going on in this, both pure protein (folding, etc), generation (RNA, etc.) and genome (timing, promoters and inhibitors, transposon controls, etc. etc.). This is all good new knowledge, which would be missed if we just sat back and said "emergent, emergent"

It is obvious that reductionism allowed very important findings in Biology.
But reduction cannot explain emergent properties arising from interactions. For example, something so "simple" as the expression of one gene cannot be only explained in terms of sequences.
In a simplistic view, reduction is centered on an static observation. There is a nucleotide sequence, there is a protein. But dynamic interactions into a cell are beyond the scope of this view.
I think that there are different levels of organization and, although the most basics be cause of the higher levels, they cannot explain them. One cell is more than the whole of its molecules.

 

[selfAdjoint]

It is obvious that reductionism allowed very important findings in Biology.
But reduction cannot explain emergent properties arising from interactions. For example, something so "simple" as the expression of one gene cannot be only explained in terms of sequences.
In a simplistic view, reduction is centered on an static observation. There is a nucleotide sequence, there is a protein. But dynamic interactions into a cell are beyond the scope of this view.
I think that there are different levels of organization and, although the most basics be cause of the higher levels, they cannot explain them. One cell is more than the whole of its molecules.

There is an unbroken causal chain from genes and proteins to cell behavior. Yes it's very complex but researchers are eagerly probing all along the chain. Your picture of oversimplified gene thought is a straw man. Nobody worth their salt in the genome project ever though single genes were the be-all and end-all of deveopment.

Way back in the stone age (1930s) both theory and experiment (with drosophilia) showed the importance of gene interactions. Transposons or jumping genes won the Nobel Prize before the genome was sequenced, and are being studied in detail now. "Gene chips" are playing a new and exciting role in studying gene interactions. I really think the cell level is the wrong level to raise the flag of emergence. But then whole organism development is not a good place to plant it either, look at the exciting work being done with yeasts, and C. Elegans (and comparisons to other caenorhabitis species). Gee I feel like a pitchman for post-genome project molecular biology, but it's a wonderful bunch of research, check it out.

 

[ryokan]

There is an unbroken causal chain from genes and proteins to cell behavior. Yes it's very complex but researchers are eagerly probing all along the chain. Your picture of oversimplified gene thought is a straw man. Nobody worth their salt in the genome project ever though single genes were the be-all and end-all of deveopment.

My view isn't oversimplified. I know from transposons and microarrays. The experiments in C.elegans are astonishing.
But I believe that your view is too deterministic. It is obvious that there are causal relationships among different complexity levels, but there is also a place for randomness, and chaos.
But my position is more related with different properties (emergent properties) at different levels. Chemistry can be explained by QM, but isn't practical to apply QM to solve all chemical problems. Is it practical the reduction of all biological questions to the most fundamental basis? Is it possible? Where is the minimum level in which we can say that the whole is more than the sum of components? Or, alternatively, do you think that this question don't make sense and all Biology could be reduced to basic Biochemistry in terms of the fundamental molecules?

 

[selfAdjoint]

Where is the minimum level in which we can say that the whole is more than the sum of components? Or, alternatively, do you think that this question don't make sense and all Biology could be reduced to basic Biochemistry in terms of the fundamental molecules?

Wherever that point may be, it is certainly not in the cells. "More than the sum of its parts" could be a pipe dream.

You mention Chemistry; do you really think that Chemistry is more than the interactions of atoms and molecules, just because it's inconvenient to calculate complex interactions ab initio? BTW, the scope and power of ab initio computing expands every year. They are now up to four component molecules I believe. Unless you believe that a thousand component molecule is somehow different in kind from a four component one, I don't see any real place for emergence in chemistry.

 

[ryokan]

Wherever that point may be, it is certainly not in the cells. "More than the sum of its parts" could be a pipe dream.

You mention Chemistry; do you really think that Chemistry is more than the interactions of atoms and molecules, just because it's inconvenient to calculate complex interactions ab initio? BTW, the scope and power of ab initio computing expands every year. They are now up to four component molecules I believe. Unless you believe that a thousand component molecule is somehow different in kind from a four component one, I don't see any real place for emergence in chemistry.

Chemistry isn't more than the interactions of atoms and molecules, of course. But, although be possible explain all the Chemistry at electronic level, it don't seems to be practical.

And Physics? From a reductionist viewpoint, how could be explained the thermodynamic's second law and the time's arrow? In this case, the high number of particles involves the need of a statistical approach. We couldn't see here an "emergent", macroscopic, non-reductible law? We pass from a time-reversible microscopic level to an irreversible macroscopic situation.

With respect to cells, it would be interesting to test the question by means of synthesis. Whereas we cannot synthetize ad integrum an eukaryotic cell from its basic molecular components, we cannot demonstrate that reduction is possible. Recently, there is effectively an interesting area on "Synthetic Biology", but it is more related to induction of changes on preexisting cells (I don't talk of viruses).

I think that we cannot discard a priori the existence of "emergent" non reductible properties in living systems, even isolated eukaryotic cells.

 

[selfAdjoint]

I think that we cannot discard a priori the existence of "emergent" non reductible properties in living systems, even isolated eukaryotic cells.

The burden is on you to show that these unobserved properties exist. You can't just wave your hands. There is as I said before detail research at every level from the whole cell down to the individual codon. Yes it's very complex. Yes they haven't nailed everything down yet. But nobody involved in this research, as far as I know, is claiming any step of the way is "emergent".

I also want to address (again) two meanings of emergent. One is, if it's convenient to redefine detail behavior in terms of larger scale behavior. Such as temperature and entropy for the momentum and energy of the micro assembly. This might be simply a convenience, a historical predjudice, or a practical matter of computation. It does not deny that EVERYTHING seen at the macro scale is caused by definite happenings at the micro scale.

The other sense of emergent is the mystical one, summed up by your phrase "more than the sum of its parts".

The first sense is perfectly scientific, and covers most of the examples you give: the chain from physics to chemistry to biology, for example. Notice that it is always assumed that future research will fill in the gaps between the levels, and in the case of biology and chemistry this has certainly happened (molecular biology), while the same thing is under way to fill in the chemistry-physics gap (ab inito calculation of molecular properties). None of this supports the mystical sense of emergence.

 

[ryokan]

I also want to address (again) two meanings of emergent. One is, if it's convenient to redefine detail behavior in terms of larger scale behavior. Such as temperature and entropy for the momentum and energy of the micro assembly. This might be simply a convenience, a historical predjudice, or a practical matter of computation. It does not deny that EVERYTHING seen at the macro scale is caused by definite happenings at the micro scale.

The other sense of emergent is the mystical one, summed up by your phrase "more than the sum of its parts".

I didn't want to use the term emergent in any mystical sense. This term and my phrase can be unfortunate, but I think that they are informative to express that the study of some macroscopic phenomena couldn't be reduced to description of components. In this respect, I posed the case of entropy. If time's arrow don't exist at microscopic level but it occurs at macroscopic level, macroscopic description cannot be reduced to mechanics of components, although this mechanics be the cause of increase of entropy. At least, from a practical viewpoint.
Yes, I agree that "everything seen at the macro scale is caused by definite happenings at the micro scale" as you said. But I doubt that description and prediction can be ever also reduced at micro scale. That would be a laplacian determinism.

 

[selfAdjoint]

I didn't want to use the term emergent in any mystical sense. This term and my phrase can be unfortunate, but I think that they are informative to express that the study of some macroscopic phenomena couldn't be reduced to description of components. In this respect, I posed the case of entropy. If time's arrow don't exist at microscopic level but it occurs at macroscopic level, macroscopic description cannot be reduced to mechanics of components, although this mechanics be the cause of increase of entropy. At least, from a practical viewpoint.
Yes, I agree that "everything seen at the macro scale is caused by definite happenings at the micro scale" as you said. But I doubt that description and prediction can be ever also reduced at micro scale. That would be a laplacian determinism.

Microscopic description of entropy is the famous work of Boltzman. After decades of controversy it is now pretty well accepted that this work was basically correct.

You seem to work the "practical"/"necessary" divide, as in your last sentence on entropy. But practical concerns in human calculation can't constrain what happens. Likewise I don't know how far to extend your Laplacian determinism remark. Do you say that because we find it difficult to compute what happens in the small, therefore Laplacian determinism doesn't exist? I don't see how that follows. On the other hand you may be saying that if we could do that computing, then we would have the powers Laplace imputed to his infinitely knowledgeable daemon. I doubt it.

 

[ryokan]

Microscopic description of entropy is the famous work of Boltzman. After decades of controversy it is now pretty well accepted that this work was basically correct.

Yes. But it remains the fact of time's arrow exists for macro, whereas there is reversibility in micro.

Likewise I don't know how far to extend your Laplacian determinism remark. Do you say that because we find it difficult to compute what happens in the small, therefore Laplacian determinism doesn't exist? I don't see how that follows. On the other hand you may be saying that if we could do that computing, then we would have the powers Laplace imputed to his infinitely knowledgeable daemon. I doubt it.

I believe that there are fundamental limits to computation.More generally, there are fundamental limits to knowledge. So, Laplacian determinism doesn''t exist.

 

[ryokan]

About the topic of this thread, I find specially interesting the papers published in Science 1999;284: 79 - 101, on complex systems, being their introduction entitled "Beyond Reductionism".