Can Math Help Predict Binding Energies in Enzymology and Crystallography?

[Bigpharma]

This is a concept that has always eluded me. I'm a Biologist who works in the Pharmaceutical Industry and I need to get a better understanding of Biophysics with respect to enzymology, specifically with regards to crystallography, electron microscopy, and catalytic site binding physics.

I Know I can go to Bio sites for the basics but I want a more mathematical description of what I see. That's why I came here.

To be real specific, I wish to find a mathematical prediction of the potential binding energies of various amino acid substitutions critical for substrate binding within the catalytic site of a reductase.

 

[cesiumfrog]

Problem is, the average physicist doesn't even know what an amino acid looks like. (Physics students memorise a minimal number of "facts".) It's too complex to solve exactly, and as far as the physics is concerned, there's nothing fundamental to learn from it.

To answer your question seemingly requires detailed knowledge of chemistry, knowledge of molecular biology, and perhaps also advanced knowledge of physics. I'm not sure whether such people are on the forum.

 

[olgranpappy]

...hmm... you probably want to talk to a quantum chemist. They love this sort of thing. What exactly do you mean by "potential binding energies?"

 

[Fra]

Sounds like a problem you might attempt to solve by first trying to write down the quantum equations describing the chemical system to some level of approximation, transform that into a well posed numerical problem that can be solved numerically and then attempt to find your answers by a high performance computer.

If you want to do it all on your own, I think you might need also some basic knowledge in implementing or using methods of computing science too.

Like olgranpappy says, that sounds just like what quantum chemists do. Trying to find the quantum properties of complex systems like larger molecules. But I think they use a lot of computing power and many techqniques to get a well posed numerical problem as analytical solutions is not even on the map.

Other biology applications are to simulated not, molecule interactions but entire organism cultures... when been attacked by interesting metabolic network modelling techniques. You define a measure to extremize (typically a condition of growth rate, and production of biosynthesis precursors etc), and use a kind of variational principle to arbitrary enzyme regulations, that are constrained to the known stochiometric matrix, making sure all the mass/energy/charge flow balancing is working. I was quite fascinated by this some time ago, and it seems the regulations magically chose by nature seem to be close to what one would expect from such methods.

Those disciplinces are often worked on by people in between computer science and biology, because the biggest problem isn't to write down the equations in absurdum, it's to get them in a numerical form that is practically solvable even on a computer. It's amazing how the computing time increases as complexity increases.

/Fredrik

 

[pkleinod]

You probably will not get the help you need from a forum unless you are very lucky. My suggestion would be to search the literature for the topic you want or, even better, look at the papers published by some well-known biophysicist who has a large group working on a variety of topics, someone like Erwin Frey, who is interested in mathematical biophysics. It is a bit of work but will, in my opinion, bring higher returns than attempting to do things by yourself. Start with google or web-of-science. Good luck.

 

[Gza]

It's quite an unfortunate fact that all too many students of Physics despise biochem, or any field related to the life sciences for that matter. The fact that there are very few -fundamental- concepts, and rather a patchwork of ideas and methodologies is the anti-motivating factor, since physicists love simplicity and elegance. The biotechnology industry is in dire need of people with the complex problem solving capabilites honed by studies in physics, since new technologies involving the reading of a genome at the single nucleotide level are beyond the grasp of people with strictly bio/chem (or both) backgrounds.

 

[Bigpharma]

It's quite an unfortunate fact that all too many students of Physics despise biochem, or any field related to the life sciences for that matter. The fact that there are very few -fundamental- concepts, and rather a patchwork of ideas and methodologies is the anti-motivating factor, since physicists love simplicity and elegance. The biotechnology industry is in dire need of people with the complex problem solving capabilites honed by studies in physics, since new technologies involving the reading of a genome at the single nucleotide level are beyond the grasp of people with strictly bio/chem (or both) backgrounds.

Very eloquently put my Friend. All the forces in the Universe are interlaced. Until we unify the disciplines of biology, chemistry, and physics such that there is now only one true Science, we will never do much more than crawl forward.

Thank you for everyone's insightful replies!

 

[Danger]

Perhaps small notices in the Biology and Chemistry forums, bringing attention to this thread, will be helpful.

 

[Bigpharma]

Perhaps small notices in the Biology and Chemistry forums, bringing attention to this thread, will be helpful.

This is a good idea. However, I confess to not knowing which thread you're referring to. I happened upon this one just surfing the internet during a moment of inspiration. As I'm sure you're aware, the internet is more rife with idiocy than not, so I was happy to find this place.

Post a link please!

 

[Danger]

By our definitions here, a 'thread' refers to this particular item that you initiated or any of those surrounding it. It's in the 'Classical Physics' sub-forum in the 'Physics' forum in the site 'Physics Help and Math Help - Physics Forums'. Just go to the main index of this site and scroll down. Right below the Engineering forum, there's one called 'Other Sciences'. Two of the sub-forums of that are 'Biology' and 'Chemistry'. It's all right here on PF.

 

[Bigpharma]

I'm sightless. Cut me some slack!

 

[Rade]

See this--contact the author at address below:

1. Quantum chemical modeling of enzyme active sites and reaction mechanisms.
Fahmi Himo* [Royal Inst. of Tech.]
Theor.Chem.Accounts., 116, 232-240 (2006)

DFT methods with B3LYP function methods and models are used to study the enzyme active sites and their reaction mechanisms using quantum chemical methods are reviewed with recent examples.

Fahmi Himo
himo@theochem.kth.se
Theor.Chem., Dept. of Biotech.
Royal Inst. of Tech.
Albanova Univ. Cent.
SE-106 91 Stockholm
Sweden

 

[Bigpharma]

See this--contact the author at address below:

1. Quantum chemical modeling of enzyme active sites and reaction mechanisms.
Fahmi Himo* [Royal Inst. of Tech.]
Theor.Chem.Accounts., 116, 232-240 (2006)

DFT methods with B3LYP function methods and models are used to study the enzyme active sites and their reaction mechanisms using quantum chemical methods are reviewed with recent examples.

Fahmi Himo
himo@theochem.kth.se
Theor.Chem., Dept. of Biotech.
Royal Inst. of Tech.
Albanova Univ. Cent.
SE-106 91 Stockholm
Sweden

Thanks Fahmi!

 

[hussness]

Hi!
I'm fairly new to this forum, but I have a background in chemistry, and have strong interests in pharmacology. I have been, over the summer, trying to find resources for mathematical biophysics. Unfortunately most people I know who do work in molecular biology or biochemistry suck at math, and most mathematicians I know don't care if anything they do might be applicable to real-world problems like this.
I had some very insightful talks with a professor in the physics department at my university this last quarter about whether or not it was feasible, given an infinite budget and a large amount of time (say 50 years) to use massively parallel systems of processors to run ab inito calculations for enzymes. He told me that he had looked into that and that the design of modern computers, even assuming Moore's law continues to hold, is limited such that that will never be possible. I wish I could express it in more formal terms or have a reference, but I don't at the moment. He said that quantum computers have shown a lot of potential for being able to solve such problems, but they are still in their infancy.

I have recently been trying to develop a basic understanding of the known motifs of protein secondary structure based on rigorous mathematical definitions. I've been wondering, for example, why phi and psi angles are used instead of a dihedral between two area vectors representing the coplanar atoms in adjacent peptide bonds. Following from this, couldn't one then use vector operators to specify a particular conformation of a biological structure?
Along this general line of reasoning, would it be possible to specify atoms and electrons with tensors (not sure if this is the correct term), basically as a data type that has certain properties, and certain operations can be carried out with these properties.
Pardon me if this has already been discussed ad nauseum, but I haven't had much luck with the search engine, and I am just starting to get into real math.

By the way, the article listed below is available for free from Springerlink. Here's the url:
Quantum chemical modeling of enzyme active sites and reaction mechanisms

 

[Bigpharma]

Everybody keeps waiting for the Pharmaceutical Industry to come up with miracle cures but the era of easy fixes is over. Only a shift in technology and perception can cure this or there will be NO SIGNIFICANT new cures in the coming century.

This necessarily entails melding the disciplines of ALL Sciences to meet a common goal.

The first person to step forward with a solution to this problem will be the world's next Bill Gates.

 

[Imagelist]

i think so

i think so .you must learn some mathe.