What attempts are there at building physical models of biological systems?

[CoffeeWLemon]

TL;DR

As title says: what attempts at building physical models of biological systems there are? Under physical model i mean attempts at describing various biological systems (for example: neurons working, some inner cell mechanist and etc) using math and physics laws. Not really important as how successful/unsuccessful or specific they are.

Hello.

I tried to give quick summery of this post in tl;dr, so here i will explain more in detail what i meaned. Also i will ask to excuse me in advince, cause i have a bad habit of expleinign something longer than it should be even after rereading and editing hull text few times and i cant really call my english good.

Quick summary on why i asking this question: i took interested in biology some time ago and where researching it in my free time. As result i come to "conclusions":

A) Living organisms, even just a cell, are extremely complex systems, to the point that we know a little how exactly they works;

B) Most biological researches, putting it simply, less "we make some mathematical model, we test this model and, if it working, we got some explanetion to procces and we can use it later" and more "we have previus experemental data, from which we can make some guesses, test it and, if we lucky, use its results". In other words, it more about result and less about understending proccess in details (which, taking i naccount complexity of researchable subjects and our interesting in using said result in medicine sooner than later, is understandable).

If take some specific example: i took some interest Yamaka researches on iPS (i interested more in inner cell workings and started to look in biology in the first case cause of that), and as i understand, it basicly (not to downplay this work or efforts, it extremly interesting data and a lot of work, with varius experements from different groups going as far as to 50s) can be sumed up as "we have pull of factors, which might have some influense on cell been in pluripotent state, so we used method of elimination to find those who really affect it", but not how exectly those factors force cell back into pluripotent state.

After having chat on this subject with people from biophysics department at university in which i studying (said department take interest in neurons working and protein binding in context of cancer), i sort of got confermation of what i got so far and came to idea that all current phycisal models of biological systems are "raw" at best. Still, i would love to get second opinion on that if possible (or even just chat on this subject), hence why i typing this.

Also, speaking a little more on what i mean under "physical models of biological systems", i guess something like perceptron model would be good example. As far as im aware (sadly i didnt have a lot time lately so extremly superficially) nowadays it used in building AI neural networks, but at first it was proposed as attempt at describing though proccess in brain. It didnt worked well as intended beside some specific cases (cause didnt coutn nonliner working of neurons, used sort of binar logic with analogy on computers and etc, if i got it right). Still, it look like interesting attempt at describing process in biological system, even with some practical use (even if not directly biological).

Just to clarify: i more interested on "what attempts" there are/were (rather than their successfulness or specificity) to get sort of idea of what we have and what we tried to use.

It would help a lot, if anyone will be able to help or even point in some dirrection, so thanks in advense.

Have a nice day everyone)

p. s. Also just to say: this post not really 100% biology, but, as far as i understand, this thread looked most suitable to post it. If it wrong place i wll repost somewhere also, sorry for inconvenience.

 

[BillTre]

First, this is the right place for this post. In my view it is 100% biology. Biology is imperialistic to other domains of thought. It goes in takes concepts and uses a lot of them for its own purposes.

Second, it would be a good idea if you would post more detail about where you got your ideas from and what kinds of applications you are interested in. Your post confused me in places and I don't really understand where you are coming from concept-wise.

There are plenty of physical models for things, especially in the past, like behavioral models of competing behavioral outputs from animal that has a choice of doing one of several different behaviors. Fight or flight, or sex and reproduction.

For modern biology as I see it, the best thing to do is build a model in your head about what is going on (for example in a cell) based on realistic ideas from research about what is known. That would be the physical model I would use for understanding a biological process. Physical models are a kind of crude way to get an idea of what you want to understand and to then build your own internal (inside your head) model for further use. The real world is always going to be much more complex.

Biophysics (which you mentioned) is full of mathematical or physical models of biological things. There are lots of similarities among flow bases processes like water flow through pipes and blood flow through blood vessels. The biological parts are much more complicated that simple plumbing however. Similarly the passage of molecules through reaction networks is somewhat like a flowing water system. I have seen passive flow of change down neuron dendrites compared to water flow in how it decreases as come of it goes out different branches.

If you have not already, you might want to take a biology course or look over a textbook. There are so many areas of biology it is hard to predict what you would want to become more informed about.
Also chemistry is important for biology. Its good to know some of that, like organic chemistry and biochemistry, but these are details.
Most biological things are highly interconnected. In addition, for each particular part in a cell, there probably many similar parts (except for the genes) doing similar things in slightly different settings in the cell. This allows them to all act different in response to something, making things even more complex

The complexity of biology is probably the main thing that sets biology apart from physics and chemistry.

 

[Andy Resnick]

TL;DR: As title says: what attempts at building physical models of biological systems there are? Under physical model i mean attempts at describing various biological systems (for example: neurons working, some inner cell mechanist and etc) using math and physics laws. Not really important as how successful/unsuccessful or specific they are.

I highly recommend checking out a few textbooks, including:

Physical Biology of the Cell (Phillips et. al)
Bioenergetics 3 (Nicholls and Ferguson)
An Introduction to Systems Biology: Design Principles of Biological CIrcuits (Alon)

 

[CoffeeWLemon]

Yes, reading it today it is in fact kinda confusing, plus "broadness" don't help at all. I will try to answer questions one at time.

In short where i got ideas of that all. I was interested in biology here and there for some time (mainly because we had biophysics department, which sounded interested, and we could get assigned to them, but in the ned i choosed theoretical physics departemnt cause of heaver math there), but were able to start some learning in my free time only somewhere half year ago (course jobs, many lessons and etc) with kinda naive idea of "it look like most health issues this way or another can be traced to "something went wrong with a cell", so why not just replace those cell? how hard it can be to "print" identical cell?". Needless to say, it was in fact hard, if not impossible, depending on what you understand under "print".

Moving on to what i already looked/read. Cause i had basicly zero knowledge on the subject at the time, beside some school programm, i looked for some textbooks to get some basis from which i would be able to start (i probably should have mentioned this in the post itself, but i knida forgot). In the end i stoped on "Life: The Science of Biology" by Sadava, Hillis, Heller, Berenbaum, mainly because there were records of lecture course on "introductory to biology" using this book on MIT site for free with all exams and problem papers. On chemistry: i had some knowledge of it already, so i just choosed some textbooks to have near me in case i didnt get/remember something. After i have done it, which took few months, and got at least some vivid ideas of what we have, i moved to pubmed to find some papers on interested subjects while using textbooks here and there (mainly because jump from "introductory to biology" to research publications is... quite big and there will be blanks to say the least).
Also, cause i dont belive there will be any better place to say it, due to some screw ups in plan of our educational program, in this semester and in next one we having biophysics (well, i mean all of us who study on this specialty) plus some chemistry was planned in the first place. I myself actually kinda happy about that, but i cant say the same for most people who join theoretical department.

As i said in post myself, i took some interest in Yamaka's researches on iPS and what we got from there. Mainly because, as i mantioned here before, i went here in the first place with interest in direct cells treatment/replacement (i guess it best word here?), and it look like the most realistic thing we have to it, where we can take cell, and transform it into another for further replacement. So answering on what "specific subject", i guess it would be inner cell working.

Right now i will kinda go a little of the way, cause i belive it would be better way to explain why i phrase it as "physical models of biological systems" at the first place and also to point about "build a model in your head about what is going on" which mister BillTre said (to get ahead, i agree with it, but i guess i see it in a bit different way). So, just to say, it defenetly gonna be sort of "projection" of how i think and see physics on biology in a way.
I will try to go with a crude analogy: lets take for example ideal gas law from physics (pV=nRT, that thing), and lets imagine here that we dont know it. We can make some experiments to find, that volume of gas expand when we heat and etc. That would be our.... sort of model/ideas of how it work on realistic experiments, which we can use to build some engine (say we heat gas and it move soem pistons). We may try to make it different depending on experiments, and we might actually upgrade it along the way, but we would be able to do it faster if we derive some, even if crude, law for it (aka pV = nRT).
So when i see something like map of different factors in cell for some specific interaction (cause building it in general, as i get it, near impossible due to how many factors there are and it just gonna be unreadable mess), i cant get away from the idea like "well yes, have idea now how they connected with each other, but if i will be doign some practical problem where i need to get one less or more, it woud be easer to work in numebrs and iwth formuls". Like taking some practical problem: one of the factors which Yamaka found have side effect that it might promote tumors grow, so we obviously trying to reduce those risk. Some experiments done on that to find soem optimal way to transport those factors and how to use them/ for how long to use them, but almost all papers which i find go way more into "we try to predictions on wide varity of subject (say mouses or in vitro) and see what will work out of them all" rather than trying to make even some crude math predictions. Clothest thing which i found, ws paper, which went into the subject of age reversal for cell in iPS, rather then repgrogramming them into pluripotent cell, and they used mathematical model of epigenetic clock to try and predict optimal window for how long we should use factors, to not transform the cell, but reverse its age.
I also should say that my anology is not the best, cause ideal gas law is... kinda way simpler than inner cell working. It almost feels like comparing "2+2" and some hgih mathematics in a way. Still, most physics branches also were not build in a day and started with some crude predictions, some took hundred years to build into something useful. So i dont really see why we should not try to give some math explanation for biological processes. Obviously they gonna be crude, most of them will not work, but giving a lot of time it might turn out into soemthign useful as well.

And going all the way back, i dont belive that only i have ideas like that. I pretty sure there are/were way more experienced people who had way more realistic experementalctations, and who tryed to do something here. I mean, in a way protein folding and perceptron model, as i see them, are something like that. So i kinda interested to see what we have already and to maybe look more into that. And to give a more specific subject, as it was suggest, soem attempts at describing inner cell working on more deep levels. If to give some example,lets say soem model, which will help us find out for how long cell must be exposed to some factor to give it a tumor, so we could be aware of it to not make it in the first place.

Anyway, i hope i were albe to clear some confusion and not make it even worser. Thank you all for your time.

 

[sbrothy]

I've wondered specifically if they've invented an electronic dog nose so it could be running constanly in airports for obvious reasons. Since they still use dogs (and I have heard even bees, however that works)i guess it's a pretty complex task. From reading online (e.g.: Towards an Electronic Dog Nose: Surface Plasmon Resonance Immunosensor for Security and Safety and Olfactory system-inspired electronic nose system using numerous low-cost homogenous and hetrogenous sensors) I can see work is being done, but again, biological systems are obviously hard to emulate.