| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 3
|
|
| Author |
|
|
GWG
Cruncher AUSTRALIA Joined: Aug 1, 2006 Post Count: 13 Status: Offline Project Badges:
|
This quote comes from the “ACM TechNews; Friday, March 9, 2007” email letter:
----------------------------------------Novel Search Engine Matches Molecules in a Flash New Scientist (03/01/07) Knight, Will Oxford University researchers have developed a way to match molecules of the same shape using the relative position of the atoms in a given molecule, an innovation that could be used to identify potential drugs very quickly. Similar shaped molecules are likely to have similar bioactive effects, causing a smaller medicinal effect in the human body. The 3D shape of a molecule is currently found by superimposing it over another molecule and measuring how much it overlaps, but this new technique, known as Ultrafast Shape Recognition (USR), maps and stores the relative position of atoms in the molecule, creating an accurate mathematical picture of its shape. The system allows molecules to be compared quickly and without concern for orientation in the database, which could prove extremely valuable given the billions of molecules currently contained in many databases. The technology could also be applied elsewhere. "Being able to match 3D objects is an important problem that might become even more important as we get more and more 3D models of our world--such as Google earth," says Carnegie Mellon image analysis expert Luis von Ahn. However, Imperial College molecular analysis expert Henry Rzepa says "part of the problem is that too little 3D shape information in a usable form actually exists. Even for molecules, proper 3D coordinates which define these shapes are all too rare." Click Here to View Full Article It is quite possible that the persons responsible for the design of the algorithms are already on to this, but if not, it occurs to me that searching a map of existing shapes and finding a close match would provide a better initial guess for shape on which to apply relaxation techniques to produce an improved shape simulation.
|
||
|
|
retsof
Former Community Advisor USA Joined: Jul 31, 2005 Post Count: 6824 Status: Offline Project Badges:
|
The research here attempts to define the shape of these protein molecules by folding and docking, since the shape isn't already known.
----------------------------------------We can't use what we don't have. Another observation is that Oxford already runs some projects on grid.org. I don't know whether this algorithm will be used in some way over there or not. It could be implemented in a future project to interact with KNOWN SHAPED MOLECULES. Grid helped out with the first phase of folding.
SUPPORT ADVISOR
Work+GPU i7 8700 12threads School i7 4770 8threads Default+GPU Ryzen 7 3700X 16threads Ryzen 7 3800X 16 threads Ryzen 9 3900X 24threads Home i7 3540M 4threads50% |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sounds like it could help with analyzing the results when we predict shapes using HPF2. It is the sort of tool that could be useful when working with our database of predicted shapes (structures).
|
||
|
|
|