Can I simulate hydrogen plus oxygen gives water products to reactants with only telling number of Hydrogen atoms and oxygen atoms where the equations satisfy scrodinger equation so the reactants will give this product for checking every chemical equations without memorizing the equations please tell me if IAM wrong simply can I check the chemical equations even the sulphuric acid preparation reactions or uranium reactions or just combine two atoms and can I tell their products
Can you please tell me what's he saing from 0:06 to 0:09, I don't get it. I mean I don't get why he concludes that "it means that they come out has hard spheres". Thanks
You somehow always are expected to assume that every phenomenon can only contain the amount of intricacy (depth, information) its investigated interactions with matter (atoms) or energy (waves) let us to consistently measure, perceive and discover. Never hypnotizing any further “weird” parameters or entities has become the norm of being smart or at least not being an idiot. Well, until you have enough proven scientific knowledge to automatically get that privilege of idiocy - given you can relate your unquestionable ultimate existing knowledge and your arbitrary idiocy.
i am using orca for QM calculations. but not able to do the MM part. is there any tutorial video for open source QM/MM interface? can anyone give any suggestion.
So how do you calculate the interactions between charged particles when there's so many of them? I'm pretty sure that calculating the attraction/repulsion between each pair of particles would be _extremely_ inefficient. Are there any better ways of doing that?
You're right that it would be extremely inefficient to treat each pair explicitly, although on a good graphics card, these kinds of simple computations can be done much more efficiently in parallel. But for a serial calculation, the clever trick performed is to use a cut-off distance. Before that distance, the pairs are calculated explicitly. Beyond it, they are smoothly switched to zero. The van der waals interactions decay as r^(-6), and are usually ok to switch off after 1-2 nanometers, or more than 10x the length of a chemical bond. With cutoffs, the efficiency of a simulation scales linearly with the system size; without cutoffs, it scales quadratically.
Oh I just remembered something else! There is a loss of information if you cutoff too soon, which happens for interactions between charged bodies using the pair-wise 'electrostatic' interaction, or Coulomb's Law which goes as r^-1. It's a bit hard to defend WHY in the comments section, but believe me that this decay is too gradual to every properly cut-off. "Everything in the universe is both gravitationally and electrostatically coupled." The clever trick that saves us this time is called a Fast Fourier Transform. Definitely beyond my ability to explain in person or in a comments section. en.wikipedia.org/wiki/Ewald_summation en.wikipedia.org/wiki/Fast_Fourier_transform
@@eriknordquist Don't worry, in my country there's a saying that pretty much translates to English as: "For a smart head, a word is enough". So all you need to do is name it, and I can find all the details myself. I'm well accustomed with FFT, so there's that, I just don't know how can this be used for calculating interactions between particles. One possibility that comes to my mind, is something similar to Taylor series approximation (which Fourier transform actually is too, since it can be represented as a complex power series): approximating the "shape" of the potential in space with lower-frequency harmonic functions where a rough approximation is OK, and adding more components where we need it more detailed. But since you mentioned the "cut off distance", there's one more thing that occurred to me: Maybe we could use Gauss's Law here in some way for those electric potentials? (it should also work for gravitation) I remember that Newton had a similar problem when he tried to calculate the gravitational attraction betwen the Earth and the Moon, and it turned out that he would have to calculate attraction between every atom of the Earth and every atom of the Moon, which would require him to calculate a six-fold integral! :q But then Gauss came in later and showed that this heinous calculation can be replaced with a simpler one (a surface integral), because you can draw an imaginary surface (a bubble) around some mass or charge density, and the gravitational attraction or electric force on its surface would be proportional to the mass or charge density enclosed inside the bubble. Which gave me an idea: Perhaps we could draw such imaginary bubbles around groups of particles and store them in a tree-like structure for faster sorting, so that when our test particle is outside a particular bubble, we wouldn't have to calculate attraction from all the bubbles inside, but they could be replaced by their common "center of mass" (or "center of charge in this case"), thanks to Gauss's law? :>
Hello there, these sources are great and thanks for sharing with us. I need to ask you something about molecular dynamics simulations which is showing in this video which program was used to create? I try to use Matlab simulation as you know that has a lot of option. Can you suggest me to use which program for this? Thanks for interesting.
That is the question of the century. Maybe they’ll come out with a decent molecular dynamics simulator and put it on steam, similar to what was done with universe sandbox.
Sorry for the very basic question, but would this method be used to predict the 3-D structure of a glycoprotein if you knew the amino acid sequence and the structure of the attached glycans?
Beautiful! Just one question - I couldn't make out the words at 0:06 - 'and what with the six twelve potentials typically used to represent such things?'
I would suppose that the x-axis is Å (10^-10 meters) and y-axis is probability of an oxygen being found at that distance from a water molecule, normalized to the average. It suggests there is structure to water; a water molecule quite frequently has oxygen atoms at a distance of 3 Å from it.
I want to do some MD simulation for interaction of surfactant and nanoparticle at oil and gas water interface. I am pretty new to this field. Can any one suggest where to start.
How do you make those simulations in your video? Is there some clever way to simulate all these atoms interacting with each other without explosion of computations?
Some simulation videos are actually timelapses sometimes. MD simulations don't really make your computer explode; it only takes mathematical calculations (which are what computers are made for anyway). It's just that it strains the computer the more variables you add to it that in needs to account for. Think of it as having to take the same simple test over millions of times in quick succession - that's stressful!
@@gelatinocyte6270 I know how computer simulations work. By "explode" I didn't mean the computer to explode, but the number of calculations to grow exponentially with every new element you add to your simulations, because the number of interactions between particles grows as well. I was rather asking about what software do they use to render such simulations and what rendering techniques do they use so that the end results look so good.
@@bonbonpony I didn't make the video, but I feel I can answer your question. There are several major simulation packages for serious simulations of biological systems, by far the most common in academic research are charmm, amber, and gromacs. Of these, gromacs is the easiest to play with without previous experience. Amber is probably fastest, and charmm is most powerful/flexible. I think gromacs can have a GUI, amber and charmm do not. You had a question about efficiency before, so I'll mention a fourth package called Desmond, which is specially written for a supercomputer called ANTON. In fact, desmond and Anton were literally made for one another, meaning they were designed together to optimize efficiency of parallel computing-- avoid the "explosions" you're rightly worried about. On Anton, a simulation that might run for weeks on a computer with a GPU, or months on a single processor, will take only MINUTES (an amazing, true fact that kind of makes me smile :). To render the movies, you can use a software called VMD, visual molecular dynamics. Both gromacs and VMD are free.
for MD simulations I use GroMaCS, and for some other 3D animations in the video you can start with blender, both are free, but gromacs runs only in Linux
i see why nano tech is so infinite and dangerous and probably exterme Fun when u have the tech and lab to practice it ...and i can see why nanomites could be a real threat..u can play God with that tech ..and thats what the most powerfull are doing i suppose ..
An "okay" job simulating water. Obviously your theory is wrong. This is what you don't have right: Do a search for "paulings omission" James McGinn Incidental Symmetry
Great explanation Thunderf00t I was half expecting you to start telling me why atoms were bullshit
thanks thunderf00t. u made this simple to understand
it’s got to be him right?
Is this Thunderf00t? amazing video mate
Thunderf00t?
Tf?😂
He really sounds like him haha
thunderfoot is that u?
Thank you - a very useful overview.
nice explanation. the simplest way to define MD simulation.
I giggle at the water jiggle @0:57
Bro wtf its thunderf00t. I'm in biochem in the year 2022. flash back boys
Garbage in garbage out! Heard the term after many years :) Thank you.
Absolutely beautiful
That's wonderful!!!
Excellent presentation. Thanks a lot!
Can I simulate hydrogen plus oxygen gives water products to reactants with only telling number of Hydrogen atoms and oxygen atoms where the equations satisfy scrodinger equation so the reactants will give this product for checking every chemical equations without memorizing the equations please tell me if IAM wrong simply can I check the chemical equations even the sulphuric acid preparation reactions or uranium reactions or just combine two atoms and can I tell their products
can anyone tell me the maximum length of time achievable for an MD run?
Duh!!
wow its very fascinating thank you.. i hope i can find more of these videos on MD
Thank you for informative lecture.
amazing video
Good video presentation.
Can you please tell me what's he saing from 0:06 to 0:09, I don't get it. I mean I don't get why he concludes that "it means that they come out has hard spheres". Thanks
Interesting video.
which software do u recommend for molecular dynamics simulations?
Thank you
Awesome.
You somehow always are expected to assume that every phenomenon can only contain the amount of intricacy (depth, information) its investigated interactions with matter (atoms) or energy (waves) let us to consistently measure, perceive and discover. Never hypnotizing any further “weird” parameters or entities has become the norm of being smart or at least not being an idiot. Well, until you have enough proven scientific knowledge to automatically get that privilege of idiocy - given you can relate your unquestionable ultimate existing knowledge and your arbitrary idiocy.
i am using orca for QM calculations. but not able to do the MM part. is there any tutorial video for open source QM/MM interface? can anyone give any suggestion.
suddenly a wild thunderfoot appeared
Nice explanation
Very informative session.
Superb!
Do these models take steric hindrance into account?
can someone help me interpret my docking results?
Nice one Thank you
i have some questions about molecular simulation, can you answer me please ?
Thanks
Nice
So how do you calculate the interactions between charged particles when there's so many of them? I'm pretty sure that calculating the attraction/repulsion between each pair of particles would be _extremely_ inefficient. Are there any better ways of doing that?
You're right that it would be extremely inefficient to treat each pair explicitly, although on a good graphics card, these kinds of simple computations can be done much more efficiently in parallel. But for a serial calculation, the clever trick performed is to use a cut-off distance. Before that distance, the pairs are calculated explicitly. Beyond it, they are smoothly switched to zero. The van der waals interactions decay as r^(-6), and are usually ok to switch off after 1-2 nanometers, or more than 10x the length of a chemical bond. With cutoffs, the efficiency of a simulation scales linearly with the system size; without cutoffs, it scales quadratically.
Oh I just remembered something else! There is a loss of information if you cutoff too soon, which happens for interactions between charged bodies using the pair-wise 'electrostatic' interaction, or Coulomb's Law which goes as r^-1. It's a bit hard to defend WHY in the comments section, but believe me that this decay is too gradual to every properly cut-off. "Everything in the universe is both gravitationally and electrostatically coupled." The clever trick that saves us this time is called a Fast Fourier Transform. Definitely beyond my ability to explain in person or in a comments section. en.wikipedia.org/wiki/Ewald_summation en.wikipedia.org/wiki/Fast_Fourier_transform
@@eriknordquist Don't worry, in my country there's a saying that pretty much translates to English as: "For a smart head, a word is enough". So all you need to do is name it, and I can find all the details myself. I'm well accustomed with FFT, so there's that, I just don't know how can this be used for calculating interactions between particles. One possibility that comes to my mind, is something similar to Taylor series approximation (which Fourier transform actually is too, since it can be represented as a complex power series): approximating the "shape" of the potential in space with lower-frequency harmonic functions where a rough approximation is OK, and adding more components where we need it more detailed.
But since you mentioned the "cut off distance", there's one more thing that occurred to me: Maybe we could use Gauss's Law here in some way for those electric potentials? (it should also work for gravitation) I remember that Newton had a similar problem when he tried to calculate the gravitational attraction betwen the Earth and the Moon, and it turned out that he would have to calculate attraction between every atom of the Earth and every atom of the Moon, which would require him to calculate a six-fold integral! :q
But then Gauss came in later and showed that this heinous calculation can be replaced with a simpler one (a surface integral), because you can draw an imaginary surface (a bubble) around some mass or charge density, and the gravitational attraction or electric force on its surface would be proportional to the mass or charge density enclosed inside the bubble.
Which gave me an idea: Perhaps we could draw such imaginary bubbles around groups of particles and store them in a tree-like structure for faster sorting, so that when our test particle is outside a particular bubble, we wouldn't have to calculate attraction from all the bubbles inside, but they could be replaced by their common "center of mass" (or "center of charge in this case"), thanks to Gauss's law? :>
Hello there, these sources are great and thanks for sharing with us. I need to ask you something about molecular dynamics simulations which is showing in this video which program was used to create? I try to use Matlab simulation as you know that has a lot of option. Can you suggest me to use which program for this? Thanks for interesting.
That is the question of the century. Maybe they’ll come out with a decent molecular dynamics simulator and put it on steam, similar to what was done with universe sandbox.
Sorry for the very basic question, but would this method be used to predict the 3-D structure of a glycoprotein if you knew the amino acid sequence and the structure of the attached glycans?
You would need ab initio homology/structural modeling for that. Something AlfaFold readily provides these days
Beautiful! Just one question - I couldn't make out the words at 0:06 - 'and what with the six twelve potentials typically used to represent such things?'
Isn't he talking about the Lennard Jones potential ?
COOL
hi, i would like to use a small part of this video for a project i am working on.
is there a way i could purchase a licence to use it?
Hi! Can someone explain what the axes are for plot at 1:32? Thanks!
I would suppose that the x-axis is Å (10^-10 meters) and y-axis is probability of an oxygen being found at that distance from a water molecule, normalized to the average. It suggests there is structure to water; a water molecule quite frequently has oxygen atoms at a distance of 3 Å from it.
Did thunder00t take this video down or something?
I want to do some MD simulation for interaction of surfactant and nanoparticle at oil and gas water interface. I am pretty new to this field. Can any one suggest where to start.
dunno if you are still interested in doing that but gromacs has amazing tutorials to simuations including 2phase interactions
ok
How do you make those simulations in your video? Is there some clever way to simulate all these atoms interacting with each other without explosion of computations?
Some simulation videos are actually timelapses sometimes. MD simulations don't really make your computer explode; it only takes mathematical calculations (which are what computers are made for anyway). It's just that it strains the computer the more variables you add to it that in needs to account for. Think of it as having to take the same simple test over millions of times in quick succession - that's stressful!
@@gelatinocyte6270 I know how computer simulations work. By "explode" I didn't mean the computer to explode, but the number of calculations to grow exponentially with every new element you add to your simulations, because the number of interactions between particles grows as well. I was rather asking about what software do they use to render such simulations and what rendering techniques do they use so that the end results look so good.
@@bonbonpony I didn't make the video, but I feel I can answer your question. There are several major simulation packages for serious simulations of biological systems, by far the most common in academic research are charmm, amber, and gromacs. Of these, gromacs is the easiest to play with without previous experience. Amber is probably fastest, and charmm is most powerful/flexible. I think gromacs can have a GUI, amber and charmm do not. You had a question about efficiency before, so I'll mention a fourth package called Desmond, which is specially written for a supercomputer called ANTON. In fact, desmond and Anton were literally made for one another, meaning they were designed together to optimize efficiency of parallel computing-- avoid the "explosions" you're rightly worried about. On Anton, a simulation that might run for weeks on a computer with a GPU, or months on a single processor, will take only MINUTES (an amazing, true fact that kind of makes me smile :). To render the movies, you can use a software called VMD, visual molecular dynamics. Both gromacs and VMD are free.
What software is this?
for MD simulations I use GroMaCS, and for some other 3D animations in the video you can start with blender, both are free, but gromacs runs only in Linux
Гальванизированный Труп Is it for Chemistry students or biology?
@cory Gromacs is definitely for molecular simulations, but Blender indeed is not.
i see why nano tech is so infinite and dangerous and probably exterme Fun when u have the tech and lab to practice it ...and i can see why nanomites could be a real threat..u can play God with that tech ..and thats what the most powerfull are doing i suppose ..
You and fer laqafxae. Poinca
An "okay" job simulating water. Obviously your theory is wrong. This is what you don't have right:
Do a search for "paulings omission" James McGinn Incidental Symmetry
Neeeeerdy. I bet narrator doesn't even have a girlfriend.
Awful! Speaker.
Nice