A new Proteine Folding project...

Message boards : POEM@HOME : A new Proteine Folding project...

To post messages, you must log in.

AuthorMessage
Profile [B^S] BOINC-SG
Avatar

Send message
Joined: 18 Oct 07
Posts: 13
Credit: 100,130
RAC: 0
Message 8 - Posted: 18 Oct 2007, 12:28:08 UTC

First of all: All the best for this new project! May there be plenty of supporters & work and may the servers be stable!

My simple question: Is this project different to the other Proteine Folding Projects?

Best, Steffen aka The Worm...


My NEW BOINC-Site

Why people joined BOINC Synergy...
ID: 8 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [AF>Slappyto] popolito

Send message
Joined: 18 Oct 07
Posts: 2
Credit: 538
RAC: 0
Message 21 - Posted: 19 Oct 2007, 5:24:57 UTC

Very good question :)
ID: 21 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile wenzel
Project developer
Project scientist
Avatar

Send message
Joined: 17 Sep 07
Posts: 168
Credit: 404,506
RAC: 0
Message 29 - Posted: 19 Oct 2007, 15:04:26 UTC
Last modified: 26 May 2008, 11:45:50 UTC

Hi,

many thanks, for the good wishes and also for joining. We are still preparing the \"science\" pages, which I hope will answer your questions thoroughly.

As a quick first answer: Protein simulation is presently dominated by two approaches.

1) You learn/copy from nature: Given a new sequence, one tries to copy structural fragments of known proteins and assembles them to good structures. Pros: Works for big proteins, gives excellent structures for high sequence similarity. Cons: Does not work, when there is no sequence similarity (new folds), does not work when there is no experimental data (transmembrane proteins, to which 40% of all known drugs bind), no kinetics

2) You simulate the folding process as it occurs in nature (Folding@Home) with a biophysical model. This is done with molecular dynamics (MD), an essentially sequential method with a time resolution of 10E-15 s/step. For a folding process in millisecond range you need A LOT of steps. Pros: full dynamics info, folding times, high accuracy structures, Cons: works only for small proteins, specific questions

POEM tries to interpolate between these two worlds. It uses an atomistic model for the protein free energy, i.e. is can work for new folds and applications in nanobiotechnology, where there is no experimental data. In contrast to MD, it exploits Anfinsons thermodynamic hypothesis (Nobel Prize in Chemistry 1972) that proteins in their biologically active state have a minimal free energy. The simulation process is thus replaced by an optimization process that is thousands of times faster than MD. Pros: Can do at least medium size proteins, gets the folding landscape, works for \"new folds\", Cons: still limited to proteins < 100 amino acids, no real kinetics (yet).

With POEM@HOME we will try to make progress on these two cons. Specifically we hope to make progress for

  • \"new fold\" proteins with low sequence similarity to existing proteins
  • proteins in nonphysiological environments (we just got a grant to develop implants, which are more biocompatible)
  • understanding the folding process of more complex proteins that cannot be studied with direct kinetic simulation
  • protein-protein interactions, where the partners change their conformation upon docking (biological signal process)
  • refinement of model structures for transmembrane proteins


I hope this helps, more later!
Cheers
Wolfgang

ID: 29 · Rating: 1 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [B^S] BOINC-SG
Avatar

Send message
Joined: 18 Oct 07
Posts: 13
Credit: 100,130
RAC: 0
Message 36 - Posted: 19 Oct 2007, 21:47:00 UTC

This explains quite a lot! Thanx! :)


My NEW BOINC-Site

Why people joined BOINC Synergy...
ID: 36 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile adrianxw
Avatar

Send message
Joined: 20 Oct 07
Posts: 88
Credit: 3,936,307
RAC: 0
Message 61 - Posted: 20 Oct 2007, 16:46:53 UTC
Last modified: 20 Oct 2007, 17:20:13 UTC

First up, best of luck with the project!

Now the questions!

When folding a novel sequence, how will your method model the action of chaperones?

Have you participated in any recent CASP competitions?
Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.

ID: 61 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile wenzel
Project developer
Project scientist
Avatar

Send message
Joined: 17 Sep 07
Posts: 168
Credit: 404,506
RAC: 0
Message 76 - Posted: 21 Oct 2007, 12:33:41 UTC - in response to Message 61.  

First up, best of luck with the project!

Now the questions!

When folding a novel sequence, how will your method model the action of chaperones?

Have you participated in any recent CASP competitions?


Great question: Chaperones act (most people believe) by lowering transition energy barriers furing folding. POEM does this during the simulation by either increasing the \"temperature\" temporaríly of by permitting the the simulation to \"tunnel\" high energy transition states.

I put temperature here in quotes, because this is not the \"system temperature\", which is defined in the entropy terms of the model, but the \"kinetic temperature\" that controls how far the simulation can go uphill.

For the \"tunneling\" we have invented the stochastic tunneling method (see PRL 1999)

For more info on the detail: Optimization on the group HP

By speeding the simulation a thousand fold (and other tricks) POEM is thousands of times faster than MD.

CASP question: Yes, we participated in CASP7. POEM is a physics based method, thus it has Coulomb interactions (O(N**2)) == EXPENSIVE. Because we did not the computational resources we did only the targets with less than 150 Amino Acids.

POEM presently uses no homology modelling. Hence we loose for template based models. For the template free models, i.e. \"new folds\" we ranked 5th (of 280 or so groups). With POEMqHOEM we hope that we get the infrastructure together to participate with the full program in CASP8.

Cheers
WW

ID: 76 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Yeti
Avatar

Send message
Joined: 17 Oct 07
Posts: 31
Credit: 139,767,077
RAC: 0
Message 82 - Posted: 21 Oct 2007, 15:56:10 UTC

So, at nect CASP, we will see a race between:

POEM@HOME
Rosetta@Home
Predictor@Home

Sounds exciting

Yeti


Supporting BOINC, a great concept !
ID: 82 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
rbpeake

Send message
Joined: 17 Oct 07
Posts: 15
Credit: 14,759
RAC: 0
Message 83 - Posted: 21 Oct 2007, 16:42:02 UTC - in response to Message 76.  
Last modified: 21 Oct 2007, 16:42:20 UTC


For more info on the detail: Optimization on the group HP

By speeding the simulation a thousand fold (and other tricks) POEM is thousands of times faster than MD.

CASP question: Yes, we participated in CASP7. POEM is a physics based method, thus it has Coulomb interactions (O(N**2)) == EXPENSIVE. Because we did not the computational resources we did only the targets with less than 150 Amino Acids.

POEM presently uses no homology modelling. Hence we loose for template based models. For the template free models, i.e. \"new folds\" we ranked 5th (of 280 or so groups). With POEMqHOEM we hope that we get the infrastructure together to participate with the full program in CASP8.

Cheers
WW


Thank you for the very interesting link, which explains the different strategies employed in the protein folding simulation problem. It is fun from my standpoint to be able to participate in these various approaches using BOINC (in the case of POEM and Rosetta@home) and through Folding@home.

The overview information is very helpful indeed, as well as your explanation of the methods you are using which take a different approach versus the other projects.

Looking forward to participating! :)

Regards,
Bob P.
ID: 83 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [AF>EDLS]_Damien
Avatar

Send message
Joined: 17 Oct 07
Posts: 2
Credit: 10,157,807
RAC: 0
Message 405 - Posted: 2 Dec 2007, 12:25:02 UTC

Please, could we know what are the diseases that Pohem@Home could especially help to cure?
ID: 405 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile wenzel
Project developer
Project scientist
Avatar

Send message
Joined: 17 Sep 07
Posts: 168
Credit: 404,506
RAC: 0
Message 407 - Posted: 2 Dec 2007, 16:26:15 UTC - in response to Message 405.  

Please, could we know what are the diseases that Pohem@Home could especially help to cure?


There are two types of projects:

1) Short term: Proteins our group is working on specifically: These include a project related to malaria and river blindness, there is a quite generic approach to cancer addressing DNA methylisation (both of these are collaborations with experimental groups). We just got funding for an experimental group dealing with drug design, which will work on chemokines. Chemokines are proteins which are involved in steering the cells of the immune system (for instance to the site of an inflammation) and POEM@HOME will support this project by modeling the interactions of chemokines with their receptors. The goal of this project is to develop molecules which block these sites and thus switch off chemokine function.

We are also working, like the rest of the planet, on beta-amyloid aggregation, which is relevant for Alzheimer and a number of other degenerative diseases of the nervous system. Quite frankly, I would not bet that computational methods will bring the cure for Alzheimer, but I think these studies are important anyways for our fundamental understanding. The other projects summarized above have a more direct implication to the related diseases.

2)Long term: The long term goal is to come up with better methods to predict protein structure, test them in CASP or other validation methods and then: make them available to biologists, pharmacists and doctors. There are many proteins in membranes (50% of all known drugs target these), but not routine methods for protein structure measurement of membrane proteins. so we presently have about 40,000 structures for proteins outside of membranes, but only 150 for proteins in membranes and many of those are closely related. Homology based methods for member in protein structure prediction on limited by the fact that there are so few structures available. So if we were able to use POEM@HOME to predict the structure of membrane proteins, we and other people could make a lot of progress for many diseases presently threatening mankind. Our contribution to this would primarily be the development of the method, rather than specific disease.

Hope this helps,
cheers
WW




Thanks for joining POEM@HOME!
ID: 407 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [AF>EDLS]_Damien
Avatar

Send message
Joined: 17 Oct 07
Posts: 2
Credit: 10,157,807
RAC: 0
Message 408 - Posted: 2 Dec 2007, 18:59:55 UTC - in response to Message 407.  

Hope this helps


It is perfect, thank you very much!

ID: 408 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
jaxom1
Avatar

Send message
Joined: 18 Oct 07
Posts: 8
Credit: 2,611,034
RAC: 0
Message 411 - Posted: 3 Dec 2007, 17:43:07 UTC

Thank you for the information.

I have added a few resources to your project today. Before, I had just signed up, but was not activly crunching.

Thanks,

Drizzt.
ID: 411 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile wenzel
Project developer
Project scientist
Avatar

Send message
Joined: 17 Sep 07
Posts: 168
Credit: 404,506
RAC: 0
Message 412 - Posted: 3 Dec 2007, 18:36:20 UTC - in response to Message 411.  

Thank you for the information.

I have added a few resources to your project today. Before, I had just signed up, but was not activly crunching.

Thanks,

Drizzt.



well, we\'re happy for everyone who joins ...

cheers
WW

Thanks for joining POEM@HOME!
ID: 412 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
staffann

Send message
Joined: 24 May 08
Posts: 1
Credit: 198
RAC: 0
Message 2157 - Posted: 24 May 2008, 15:13:59 UTC - in response to Message 29.  

It seems that we have a difference of opinion regarding POEM vs Rosetta:

I am trying to understand the difference between the Rosetta and the POEM@home projects. In the POEM forum, I found a post that explains it like this:

As a quick first answer: Protein simulation is presently dominated by two approaches.

1) You learn/copy from nature (ROSETTA@HOME is a good example). Given a new sequence, one tries to copy structural fragments of known proteins and assembles them to good structures. Pros: Works for big proteins, gives excellent structures for high sequence similarity. Cons: Does not work, when there is no sequence similarity (new folds), does not work when there is no experimental data (transmembrane proteins, to which 40% of all known drugs bind), no kinetics

2) You simulate the folding process as it occurs in nature (Folding@Home) with a biophysical model. This is done with molecular dynamics (MD), an essentially sequential method with a time resolution of 10E-15 s/step. For a folding process in millisecond range you need A LOT of steps. Pros: full dynamics info, folding times, high accuracy structures, Cons: works only for small proteins, specific questions

POEM tries to interpolate between these two worlds. It uses an atomistic model for the protein free energy, i.e. is can work for new folds and applications in nanobiotechnology, where there is no experimental data. In contrast to MD, it exploits Anfinsons thermodynamic hypothesis (Nobel Prize in Chemistry 1972) that proteins in their biologically active state have a minimal free energy. The simulation process is thus replaced by an optimization process that is thousands of times faster than MD. Pros: Can do at least medium size proteins, gets the folding landscape, works for \"new folds\", Cons: still limited to proteins < 100 amino acids, no real kinetics (yet).

With POEM@HOME we will try to make progress on these two cons. Specifically we hope to make progress for


* \"new fold\" proteins with low sequence similarity to existing proteins
* proteins in nonphysiological environments (we just got a grant to develop implants, which are more biocompatible)
* understanding the folding process of more complex proteins that cannot be studied with direct kinetic simulation
* protein-protein interactions, where the partners change their conformation upon docking (biological signal process)
* refinement of model structures for transmembrane proteins



Is this a correct explanation, especially regarding Rosetta? The statement that Rosetta doesn\'t work for new folds surprised me a bit.





It is wrong. Rosetta uses a physically detailed all atom model, and has been unquestionably the best method to date for new folds, as well as for designing new proteins. the person who wrote that text is clearly not very well informed about the results of the CASP experiments or recent progress in protein design.

The last answer is from Dr Baker himself. I really would like to understand the difference between the projects, but the above makes that hard. Any comments?
ID: 2157 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sid

Send message
Joined: 12 Dec 07
Posts: 16
Credit: 17,681,757
RAC: 0
Message 2169 - Posted: 25 May 2008, 14:27:37 UTC




ID: 2169 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile wenzel
Project developer
Project scientist
Avatar

Send message
Joined: 17 Sep 07
Posts: 168
Credit: 404,506
RAC: 0
Message 2182 - Posted: 26 May 2008, 11:40:33 UTC - in response to Message 2157.  
Last modified: 26 May 2008, 11:46:14 UTC

Rosetta uses a physically detailed all atom model, and has been unquestionably the best method to date for new folds, as well as for designing new proteins.


It is true that Rosetta uses an all-atom forcefield, although this forcefield has, to my knowledge, not been used to fold proteins from extendend conformations by emulating the folding pathway.

I would also agree that Rosetta is among the best methods for prediction of new folds to date, but the results of the recent CASP experiments show that there is much room for improvement for such targets. The fact that new fold predictions are very difficult and that the quality of the present predictions falls far behind experimental methods is undisputed. This is not a criticism of Rosetta, but simply the state-of-the-art in the field. Presently all available methods give only occasionally accurate structures, it is difficult to predict beforehand, when such a success occurs and the size of the proteins for which such successes have been demonstrated remains limited.

The working hypothesis of our approach is that a biophysical forcefield that could fold many proteins from completely random conformations is a good candidate to rank-order structures for protein structure prediction. Such methods were initially believed to contribute significantly to protein structure prediction in the first CASP rounds but have now completely taken the backseat to homology/fragment based methods.

This is what we are presently testing in our CASP participation.

Cheers
WW

PS: The section starting:
1) You learn/copy from nature ....

in the write-up refers to template based modelling and not specifically to Rosetta (which was just cited as one of the most important fragment based methods)
Thanks for joining POEM@HOME!
ID: 2182 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : POEM@HOME : A new Proteine Folding project...


Copyright © 2017 KIT-INT