Starting Point of Simulation?

Moderators: Site Moderators, FAHC Science Team

Post Reply
Jonathan
Posts: 23
Joined: Fri Jul 08, 2011 4:52 pm

Starting Point of Simulation?

Post by Jonathan »

So you guys do not use xray-crystal as I understand from a previous post. What is the starting point of a protein? Is it in its primary structure.
Jesse_V
Site Moderator
Posts: 2851
Joined: Mon Jul 18, 2011 4:44 am
Hardware configuration: OS: Windows 10, Kubuntu 19.04
CPU: i7-6700k
GPU: GTX 970, GTX 1080 TI
RAM: 24 GB DDR4
Location: Western Washington

Re: Starting Point of Simulation?

Post by Jesse_V »

I think that no matter what conformational shape the protein is in, its primary structure will always be the same since that's its amino acid sequence. There's probably not a uniform shape that it starts from, and they can start from whatever shape they'd like since it's all virtual. Interestingly, F@h doesn't seem to start from the beginning and run really long simulations all the way to a particular final state. It's a preliminary summary and needs some revising and further details, but based on a F@h paper I made an explanation of their methods here http://en.m.wikipedia.org/w/index.php?t ... gnificance that you might find useful in answering this question.
F@h is now the top computing platform on the planet and nothing unites people like a dedicated fight against a common enemy. This virus affects all of us. Lets end it together.
Jonathan
Posts: 23
Joined: Fri Jul 08, 2011 4:52 pm

Re: Starting Point of Simulation?

Post by Jonathan »

Thanks for your response, I don't think I was clear.

I was asking how do they build the .pdb which tells the proteins information in 3D (X,Y,Z Coordinates) . Yes the primary structure is a straight line of aa's, but i was asking IF that is what they do to start off the simulation.

As for you saying that the starting shape doesn't matter, I do not think that is correct. If you played with FoldIt or used OpenMM the starting point of simulation is a big factor of what shape it comes to be. Maybe this doesn't matter because of Markov states but even then, it should have a time factor involved (i.e. one starting point being quicker than another starting point).
Jesse_V
Site Moderator
Posts: 2851
Joined: Mon Jul 18, 2011 4:44 am
Hardware configuration: OS: Windows 10, Kubuntu 19.04
CPU: i7-6700k
GPU: GTX 970, GTX 1080 TI
RAM: 24 GB DDR4
Location: Western Washington

Re: Starting Point of Simulation?

Post by Jesse_V »

No I never said the starting states are equal and picking one over the other doesn't matter, I was saying that it's possible for them to start a simulation from any conformation they want. In this way, my point was that there's probably not one universal start. There would be a time value involved, or at least a probability of the protein moving from that state and eventually end up in another given state. MSMs are pretty statistical, there's a lot of "what's the chance that the protein will follow this route? Could it change and take another folding pathway?" going on. So picking a given state could have some influence, but then again based on how the MSM is set up its statistically possible for two different starting states to end the same way. At least that's my understanding. I have no idea how they build the file though.
F@h is now the top computing platform on the planet and nothing unites people like a dedicated fight against a common enemy. This virus affects all of us. Lets end it together.
gwildperson
Posts: 450
Joined: Tue Dec 04, 2007 8:36 pm

Re: Starting Point of Simulation?

Post by gwildperson »

To understand the complete MSM model, you have to gather the statistics of all possible confirmations and understand the probabilites of each possible state morphing C1 C3 C3... whether you have simulations that start at state B or only the simulations that start at A and move to B are considered.
Stonecold
Posts: 332
Joined: Sun Dec 25, 2011 9:20 pm

Re: Starting Point of Simulation?

Post by Stonecold »

Jesse_V wrote:I think that no matter what conformational shape the protein is in, its primary structure will always be the same since that's its amino acid sequence. There's probably not a uniform shape that it starts from, and they can start from whatever shape they'd like since it's all virtual. Interestingly, F@h doesn't seem to start from the beginning and run really long simulations all the way to a particular final state. It's a preliminary summary and needs some revising and further details, but based on a F@h paper I made an explanation of their methods here http://en.m.wikipedia.org/w/index.php?t ... gnificance that you might find useful in answering this question.
Um, you posted the link to the mobile version of Wikipedia... It gave me quite a surprise, actually. I thought Wikipedia had changed their layout! :shock:
Jesse_V
Site Moderator
Posts: 2851
Joined: Mon Jul 18, 2011 4:44 am
Hardware configuration: OS: Windows 10, Kubuntu 19.04
CPU: i7-6700k
GPU: GTX 970, GTX 1080 TI
RAM: 24 GB DDR4
Location: Western Washington

Re: Starting Point of Simulation?

Post by Jesse_V »

Stonecold wrote:Um, you posted the link to the mobile version of Wikipedia... It gave me quite a surprise, actually. I thought Wikipedia had changed their layout! :shock:
Oops! Apologizes, my mistake. Yes I imagine that could be startling! That must have slipped through, normally I try to delete the mobile parts of the URL. My laptop fan hasn't been acting right recently, hence the mobile version.
F@h is now the top computing platform on the planet and nothing unites people like a dedicated fight against a common enemy. This virus affects all of us. Lets end it together.
Stonecold
Posts: 332
Joined: Sun Dec 25, 2011 9:20 pm

Re: Starting Point of Simulation?

Post by Stonecold »

You could use the Maxthon browser. It's extremely fast but doesn't tax the CPU or generate much heat.
Jesse_V
Site Moderator
Posts: 2851
Joined: Mon Jul 18, 2011 4:44 am
Hardware configuration: OS: Windows 10, Kubuntu 19.04
CPU: i7-6700k
GPU: GTX 970, GTX 1080 TI
RAM: 24 GB DDR4
Location: Western Washington

Re: Starting Point of Simulation?

Post by Jesse_V »

Stonecold wrote:You could use the Maxthon browser. It's extremely fast but doesn't tax the CPU or generate much heat.
Hmm. I've never heard of that browser before. I normally run Chrome, but Maxthon probably won't make much differrnce considering that I'm fairly sure the bearings in the fan are bad or something, since even at total idle it makes a lot of noise and vibrates the whole machine. Normally it operates as a soft hum, even when SMP+GPU folding.
F@h is now the top computing platform on the planet and nothing unites people like a dedicated fight against a common enemy. This virus affects all of us. Lets end it together.
Post Reply