On Thursday and Friday last week I had the privelage of attending Biological Complexity: From Molecules to Systems. There were many very interesting talks from members of the Weizmann Institute as well as by researchers at UCL and other UK institutions.
The experience underlined the idea that there are people modelling just about everything; and over the next few blog posts I will give you some examples that I thought were elegant (or that pleased me for some other reason).
Today, though I would like draw your attention to the special online collection of articles at Science entitled Forests in Flux mentioned in the keynote talk given by Stephen Emmott from Microsoft Research, Cambridge.
There are also two videos of Drew Purves talking about modelling forest dynamics that you might find interesting.
Link to the Climate Feedback blog.
Interesting topic, I was wondering just yesterday: Is there a good definition for a ‘system’? Reason why I am asking, I am used to people calling everything a ‘system’ that is described by their model, without any specification of exactly what type of thing you’re trying to describe. E.g. I would have called a molecule a ‘system’ too. It maybe doesn’t have some characteristics that an organism has, e.g. the organism is certainly a more complex system, and has some other properties as well, but I’d have thought also a molecule is some sort of ‘system’ in that you have a set with elements that stand in some relation to each other, and so on.
?
I also keep hearing about systems biology and there seems to be more than one definition out there. I guess that the system comes from the fact that there is a collection of units that puts together serves some sort of function?
While there are some topics that quite obviously fall under the heading of “systems biology”, I am still (even after 2 days of listening to talks) not sure exactly where the boundaries are. And I agree that for each system, a precise definition at the outset should be a requirement as it is a word with so many meanings.
Stephen Emmott said something like:
“We have spend the last 50 years pulling things apart (creating lists of parts, lists of reaction, lists of interactions), and we are going to need to spend the next 50 years putting them back together… and this is what we are calling ‘Systems Biology’.”
(these were not the exact words, but hopefully I have conveyed the idea that he was trying to communicate)
Yes, i think the “emergent properties” aspect, as well as the “taking apart and putting together” is crucial.
Hi There,
Sorry for being dumb, but I’m really confused about the question what a system is. Maybe this becomes clearer when I ask you what is NOT a system.
Maxine, my problem is that the properties you define I would have attached to a certain class of systems: complex systems. Complex systems how I am used to the term are those in which the whole has properties the constituents (possibly: subsystems) don’t have – there is essential information in how the parts work together, and it has emergent features.
But you can have other types of systems. Natural systems, open or closed systems, steady state systems. And so on. But what is a system without any specification?
Now when I try to find something that is NOT a system the only examples that I can come up with are mathematically abstract things. That is certainly possible, that you have a set of elements without any relation whatsoever. But leaving aside mathematics and speaking of the real world, I come to think everything is a system. It might be a simple system, or a boring system. A static system or a highly symmetric one. But what is not a system? An atom is a system. A single particle is also a system as far as I am concerned. It seems to be a rather trivial one, possibly, until you realize it has a field around it, that itself consists of infinitely many virtual particles.
So I am left to wonder whether the ‘System’ in the title of that symposium is actually short for ‘natural system’ or ‘complex system’ or ‘self-sustaining reproductive system’ or whatever, because just as it stands, I can’t make sense of it.
Sorry if that sounds like nitpicking. The reason why I am confused is that I would have referred to anything that a model describes as ‘the system’, and I am wondering how that relates to the use of the word in other contexts. In my field, the universe is a ‘system’ as much as two photons are.
Best,
B.
Sabine: I am definitely with you in that “system” has too many meanings for it to be intuitive what is meant by many titles with the word “system” in them. And I agree that the authors that use this word should define their system as quickly and unambiguously as possible. (The title of the blogpost was a bit silly really as neither idea was developed in the text. Perhaps “Bronwen went to some talks last week” would have been better.) :)
I think, though, that we can limit outselves slightly by adding the word Biology to make “Systems Biology”. For the talks that I attended, the spectrum started with protein domains and ended with forests, and I am not sure exactly where we passed from the “Molecule” to the “System”. I would like to think that systematically organising the thousands of structural domain possibilities should count as “Systems Biology”, but I have a feeling that that would be open to debate.
I think that we could say that the set “Systems Biology” contains only “complex systems” (as defined by Sabine and Maxine: Complex systems are those in which the whole has properties the constituents (possibly: subsystems) don’t have).
Please say if you disagree on this point! :)
In the course of the talks, the Solar System was mentioned for illustrative purposes and some of the ideas used to model biological system seem to borrowed from optimising computer systems or “mechanical systems” (like cars). So even if the there is a firm boundary to what we call “Systems Biology”, the people who are doing it do not seem to be bounded to biology in their thinking.
Bit of a novice here, but I thought the definition of a “system” was actually rather well defined in this context .ie. a “system” is the idealised, self-contained subject of the model. That is, everything that is required to successfully model the system has been taken into account.
For example, a lot of cells can be modelled as self-contained systems. If a crucial external hormone interaction is required to complete the model but is not included in that model, then its not really a system and hence your model will not be a success. However, these external influences can be dealt with up-front with constants or included in the model itself to re-instate the “system” definition.
In reality, nothing is so idealised. Even ab initio calculations used in theoretical chemistry only work (say by calculating the structure of some impossible molecule) by reducing the complexity of the electronic and matrix interactions far as possible.
Hi Bronwen,
Thanks for the clarification, I think I agree with you. I didn’t mean to criticize the title of this post, it was just a coincidence I spend the weekend with Lazlo’s “Systems View of the World” and though the spirit resonates with me, I couldn’t make much sense of the concept.
Anybody here read Lazlo’s book?
Having given it some more thought, I think what puzzled me is that the book (and similar stuff you find online about ‘General Systems Theory’ and ‘System Science’) is said to be a ‘paradigm shift’, that I think goes against the paradigm of reductionism. Being left to wonder what the alleged shift is, I guess it must have passed me by, and possibly was before my time? I can recall one of my profs proclaiming repeatedly physics is the ‘queen’ of sciences (the German word for physics is female), which meant to say it’s the must fundamental science in that it talks about the smallest constituents. I never understood though why this makes it a better science than others where the constituents are not ‘fundamental’ but possibly subsystems. This I guess is the essence behin the ‘systems view’ – that you don’t resolve the constituents but look for the dynamics of the whole as it is.
Best,
B.
Hi Alain,
Bit of a novice here, but I thought the definition of a “system” was actually rather well defined in this context .ie. a “system” is the idealised, self-contained subject of the model. That is, everything that is required to successfully model the system has been taken into account.
Now you’ve confused me again. I think you are mixing up ‘system’ with ‘model’. Could you clarify what you mean with either. I would agree that the system is the subject of the model, but that what is idealized and simplified and so on, I’d have said is the model of the system. Also, I can’t make sense out of the last sentence: has to be taken into account… for what?
Best,
B.