Folding Forum

Hello everyone!

After reading project 7809's description, and subsequently, Christian's "Project Manager Bio", I am now slightly confused on protein folding simulation accuracy. What brought upon this confusion is when it says how

he is most interested in how protein folding in a cell is different than folding in a typical experiment or simulation

and that

The conditions are very different when comparing in vivo (i.e. in a cell) to in vitro (i.e. in a test tube) folding. For example:

1) Protein folding can occur co-translationally. This means that the protein can fold as it is being synthesized

2) Proteins in the cell are surrounded by a lot of things, like salts and chaperones, which can have a profound effect on the folding process

For these reasons, Christian would like to explore the folding process in the cell, which can hopefully provide new insight into the protein folding problem.

I realize that the simulations other than his that are being run must be accurate enough that they are worth performing and having papers written about. I would just like to know how much of an effect and how much different the conditions mentioned above make on the accuracy of simulations that do not take those conditions into consideration.

Hopefully this does not require a very scientific answer lol, or, if it does, that it can be dumbed down enough for the average person to understand. Thank you!

Good question! One F@h paper that came to my mind that I think you should check out is "Protein folding under confinement: A role for solvent". You can read it here: http://www.pnas.org/content/104/25/10430.full I think it will provide more information that you might find useful to help answer this question. Let me know if you need help understanding it or something. I hope most everything in that paper is understandable to you, though you can skip the "Results and Discussion" and "Methods" sections since they're entirely biochemistry technobabble.

Protein chaperones help a protein fold in a crowded molecular environment, and try to keep it from messing up completely, so they are a pretty important piece for cellular survival.

Ah thank you for replying! I don't quite understand what that paper was saying to be honest lol, but that's coming from one with no biochemistry background.

My first question is would it be safe to say that projects like 7809 that focus on folding with many variables, as stated in the paper (at least from what I got), are significantly more scientifically accurate for their overall purpose because of the avoidance of things that can be missed by not having many variables? Or are they not more accurate for their overall goal because, due to their complexity, there are many things that they must specifically focus on and are therefore missing a lot of possibilities? It just seems to me that projects that do not take into account all the variables can be very inaccurate. Right now my opinion is that all projects should be like Christian's because of the apparent vast difference in results with certain confinements associated with in vivo

The mathematical model of a protein that is used to simulate folding is always an approximation. It's important to include things that might change the answer and to exclude things are are not important. So just what is important?

The fundamental question being asked both by this research (and by you) is just how much of the environment which a disease is able to invade must be present to get a valid picture of how the disease works. Do we need the entire human body? Do we need an entire cell? Do we need other proteins to be nearby? Do we even need a solvent to simulate the fluid around the protein? Do we need the entire protein, or can we learn something by simulating a part of a protein?

Simulating a part of a protein is a challenging mathematical problem probably requiring the resources of FAH for many weeks. As the material being simulated grows larger and larger, more and more computer resources are needed. (An entire cell is essentially impossible, within a lifetime.) Thus the scientific challenge it to make sure that the simulation covers enough AND that it can be achieved within a feasible time-frame.

There are other questions, of course, like how complicated an equation do they need to use for the forces and how fine a time resolution do they need to calculate but they all boil down a knowing when enough detail has been included and yet avoiding too much detail.

Hopefully his study will tell the scientific community more than they know now about how to conduct future studies.

FAH has already won awards for simulation accuracy vs experimental results. So to me, this project, in simpler terms, looks to be a simulation with more detailed results. Or like taking a picture with a higher resolution so you can see the finer details.

Thanks! I think I understand. The main reason I was biased toward this project was this paragraph from the paper Jesse referenced (which I forgot to post last post):

Recent folding simulations of purely polymeric models and models that treat solvent explicitly have shown drastically different results. Specifically, Ziv et al. (10) have suggested that for a small helical peptide helix formation is stabilized upon confinement to a cylindrical cavity. These results were explained in terms of polymer entropy arguments as described above. On the other hand, Sorin and Pande (17) have recently shown that for an α-helical peptide confined to a single-walled carbon nanotube with explicit solvent the opposite effect was observed; the unfolded state is stabilized and the helix unfolds. This observation was explained in terms of solvent entropy. In bulk, protein folding maximizes solvent entropy, but in a confined system solvent entropy is already limited and protein–protein interactions experience a reduced entropic stabilization relative to protein–water interactions.

So even though some other projects don't take into account all the things this project does, all data is good data, and a quicker result must trade with a loss in complexity and scope.

Maybe I just shouldn't worry too much about what is and isn't accurate for the research since there are much smarter people taking care of it that actually know what they are doing and talking about

Wanting to know that your contributions are going towards scientific studies that are applicable and helpful is a legitimate question/concern. I hope you'll consider the question answered.

I do! Thanks to Tim, Jesse, and Bruce!