Tim Duignan's banner

Tim Duignan

@TimothyDuignan • 6,251 subscribers

Modelling and simulation using quantum chemistry, stat mech and neural network potentials #compchem #theochem at Orbital

Shorts

Running out of memory used to be a common headache when running molecular simulations with neural network potentials. Not any more. Here Orb is simulating over half a million atoms on a single GPU (H200). This is a fully solvated COVID spike protein. (The trick is to specify brute_force_knn=False in the calculator.) This is more than an order of magnitude larger than some other universal models. I can get over 7,000 steps of molecular dynamics per day with it.

Running out of memory used to be a common headache when running molecular simulations with neural network potentials. Not any more. Here Orb is simulating over half a million atoms on a single GPU (H200). This is a fully solvated COVID spike protein. (The trick is to specify brute_force_knn=False in the calculator.) This is more than an order of magnitude larger than some other universal models. I can get over 7,000 steps of molecular dynamics per day with it.

106,545 Aufrufe

So I think I've found another pretty incredible example of the generalisability of neural network potentials: this is a problem I've been dreaming of tackling for a decade but never felt I had the the tools to get at until now: How do potassium ion channels work. 1/n These channels are ubiquitous throughout biology they carry potassium ions out of cells with incredible speed and selectivity and are critical for nerve functioning and neuronal firing etc. Despite this we still don't know basic things about them. In fact, debate has gone on for years in top journals about whether water is co-transported with potassium or not.

So I think I've found another pretty incredible example of the generalisability of neural network potentials: this is a problem I've been dreaming of tackling for a decade but never felt I had the the tools to get at until now: How do potassium ion channels work. 1/n These channels are ubiquitous throughout biology they carry potassium ions out of cells with incredible speed and selectivity and are critical for nerve functioning and neuronal firing etc. Despite this we still don't know basic things about them. In fact, debate has gone on for years in top journals about whether water is co-transported with potassium or not.

95,020 Aufrufe

Here's a fun one you can't do in the lab: diamond melting at thousands of degrees simulated with Orb. You can simulate anything you want with it here:

Here's a fun one you can't do in the lab: diamond melting at thousands of degrees simulated with Orb. You can simulate anything you want with it here:

31,130 Aufrufe

So Orb has blown me away again. I simulated the carbonic anhydrase enzyme with it: one of the most important and well studied enzymes in biology. (It converts CO2 to bicarbonate and is involved in many diseases and could also be useful for carbon capture.) Remarkably, despite only being trained on small crystals, it manages to spontaneously reproduce the proton relay carrying protons from the zinc-bound water to the His64 residue — a mechanism that took over a decade of careful experimental studies to unravel. Even more surprisingly, it also suggests a new previously unreported reaction mechanism for forming bicarbonate, which actually makes a lot of sense and independent quantum chemistry show is plausible. (Done with Rowan). Preprint here: You can check out the trajectories yourself, i.e,: (Thanks Alexander Mathiasen) Orb itself can be found here:

So Orb has blown me away again. I simulated the carbonic anhydrase enzyme with it: one of the most important and well studied enzymes in biology. (It converts CO2 to bicarbonate and is involved in many diseases and could also be useful for carbon capture.) Remarkably, despite only being trained on small crystals, it manages to spontaneously reproduce the proton relay carrying protons from the zinc-bound water to the His64 residue — a mechanism that took over a decade of careful experimental studies to unravel. Even more surprisingly, it also suggests a new previously unreported reaction mechanism for forming bicarbonate, which actually makes a lot of sense and independent quantum chemistry show is plausible. (Done with Rowan). Preprint here: You can check out the trajectories yourself, i.e,: (Thanks Alexander Mathiasen) Orb itself can be found here:

22,643 Aufrufe

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

This is the most surprising and exciting result of my career: we were running simulations of NaCl with a neural network potential that implicitly accounts for the effect of the water, ie a continuum solvent model (trained on normal MD) when Junji noticed something strange: 1/n

This is the most surprising and exciting result of my career: we were running simulations of NaCl with a neural network potential that implicitly accounts for the effect of the water, ie a continuum solvent model (trained on normal MD) when Junji noticed something strange: 1/n

1,940,398 Aufrufe • vor 2 Jahren

Uranium is one of the hardest elements to simulate due to the large number of electrons, so I thought I would try it with Orb and remarkably it seems to behave well, even getting the melting point roughly correct. I think looking at this systematically for many metals with Orb would be a great research project.

Uranium is one of the hardest elements to simulate due to the large number of electrons, so I thought I would try it with Orb and remarkably it seems to behave well, even getting the melting point roughly correct. I think looking at this systematically for many metals with Orb would be a great research project.

10,668 Aufrufe • vor 1 Jahr

Keine weiteren Inhalte verfügbar