The Birth of Quantum Biology

Roland Piquepaille writes "Just when you finally have grasped the concept of quantum mechanics, it's time to wake up and to see the arrival of a nascent field named quantum biology. This is the scientific study of biological processes in terms of quantum mechanics and it uses today's high-performance computers to precisely model these processes. And this is what researchers at Rensselaer Polytechnic Institute (RPI) are doing, using powerful computer models to reveal biological mechanisms. Right now, they're working on a "nanoswitch" that might be used for a variety of applications, such as targeted drug delivery to sensors."

Comments

Comment removed

[account_deleted]

Comment removed based on user account deletion

How is this any different?

[dorpus]

Scientists have been building 3-D computer models of organic molecules since at least the 1980s, using the same equations to predict likely reactions. It sounds like plain biochemistry given a new window dressing.

Re:How is this any different?

[wass]

Scientists have been modelling chemical systems within the quantum realm for almost a century now. The problem is that there are very few problems which can be exactly solved. Eg, the hydrogen atom is one of the few solvable ones, but in reality that's only solvable when ignoring all the fine structure corrections (no spin-orbit, relativistic, or spin-spin perturbations). Once you get to the 'difficult' problem of only a mere helium atom, which in its simplest form neglecting fine structure is 'only' two interacting electrons orbiting a nucleus that you model as just a point mass with charge +2e, things get very complicated very quickly. Now imagine modelling something more complicated like a benzen ring, then imagine an actual protein.

This isn't anything new per se, just that the complexity of the modelled systems is getting larger, and due to the numercal estimation processes needed to get anything remotely usable these realms haven't been accessible until lately with the increase of computing power. So where does one draw the line between physics, chemistry, biochemistry, and biology? In these cases, what's being modelled are primarily systems consisting of electrons, neutrons, and protons, interacting with Coulomb force (like-charges repel), spin-orbit interactions, spin-spin interactions, Pauli-exclusion principle, etc. Add more atoms, system gets more complicated, and needs bigger computers.

So it's an age-old problem, using almost age-old numerical techniques, running on new shiny computing clusters

Overhyped

[AFairlyNormalPerson]

"This is the scientific study of biological processes in terms of quantum mechanics and it uses today's high-performance computers to precisely model these processes."

Precisely modeling these processes? Biggest overstatement EVER. Total hype.
When looking at large systems you are screwed and you can generally screw yourself in 1 of 2 ways:
1) Preciesly model few configurations, in which case, your results are not comparable to reality, which is an ensamble average over billions of configurations
2) Model things in an emprirical/semi-empirical, yet surprisingly CRUDE way: allowing one to sufficiently sample phase space, but not in an analytically useful way.

Quantum mechanics in biological systems are typically done with QM/MM, where the "QM" is semi-EMPIRICAL, i.e., it takes parameters. These methods and parameters were NOT designed with biological systems in mind. They were chosen to reproduce small molecule heats of formation. People have found that they work poorly for biological studies unless they are reparametrized (quite frankly, you need to know "the answer" in order to get "the answer" "right") or unless other post-priori, ad hoc corrections are applied. Only a small portion of people who use QM/MM actually reparatrize the semiempirical method and those who do find the new parameters are not very transferable for use between different types of biological systems. For crying out loud, most semiempirical hamiltonians don't even provide the functional forms needed for some of the most basic molecular interactions, e.g., London dispersion, proper polarization to external fields, hydrogen bonding, orthogonalization errors in torsional barriers, etc..

This stuff isn't really new and it's extremely overhyped.

Re:Overhyped

[Anonymous Coward]

I totally agree. I am a computational chemist who uses QM/MM modelling of biological systems and the field is definitely not new. It is an exciting field of scientific study with a lot of promise, and I do commend the scientists involved for popularising their research. The best thing about this type of science is that it combines the best parts of chemistry, physics, maths and computer science. In my opinion, the biggest challenge at the moment is how do we (a) increase the detail (expense) of the QM calculation, so that the energy of the biological molecule is calculated accurately, while (b) reducing the computational expense of the calculation so that billions of configurations of the molecule can be evaluated so that we can get a handle on the entropy. Given that we are dealing with tightly coupled equations that scale with N^5 and above, just buying more and bigger computers is not a solution... ;-)

Re:Overhyped

[Anonymous Coward]

The beauty of QM/MM simulations is that you can use MM to find the relatively few conformations that are worthy of full QM analysis. MM simulations do a pretty good job of predicting biologically relevant conformers. With that information you can then do full blown QM simulations on selected conformers to predict/model covalent chemistry. This approach has been particularly successful at modeling hydrogen tunneling (see the works of Prof. Truhlar at the U. of Minnesota). Suggesting that current efforts are irrelevant because you can't model all possible conformers in QM detail is just plain inaccurate.

Re:Did someone say Quantum Biology?

[JabberWokky]

As a quantum chemist, my wife (and all associates) tend to prefer the term "quantum mechanics" rather than "quantum physics". I've noticed that seems to be the term used in papers anyway.

.

I also wonder how this is at all new... she models inter-molecular protein reactions using high speed computers and the field has been doing so for quite awhile. The code is in Fortran77, as that seems to be the popular language for such research. It's not that it's not an interesting field, it's just not really a "nascent field" (at least as described by the term "using powerful computer models to reveal biological mechanisms"). news.rpi.edu, alas, appears to be suffering right now, and nobody has posted a mirror.

--
Evan