In addition to protein structure prediction

Quoting from https://boinc.bakerlab.org/rosetta/rah_about.php
"Rosetta@home is a scientific research project that uses internet-connected computers to predict and design protein structures, and protein-protein and protein-ligand interactions."

So, wrt protein-ligand functions of Rosetta, what would be the differences and similarities with software such as AutoDOCK (used by FightAIDS project) and LigandFit (used by the CureCancer project), or the THINK software prior to that.

Could the Rosetta software be used in such projects in the future?

Also interesting on the same issue is a recent dialogue in F@H's forums:

The current industrial standard of the computational drug design is the virtual screening using the chemical library and a proven program called DOCK. As far as I am aware of, I have never seen any of the result from FAH actually used in the drug development. Maybe the FAH ppl can correct me on this. -mshen


We have a big push to develop methods much more accurate than DOCKing. DOCKing has been shown to be useful for many applications in drug design, but not very good at predicting affinities (which is what's really important for knowing how well a drug binds). V.Pande, FAH head scientist

I know (as well as everyone else in the field of computational biology) your group spent a lot of efforts on calculating the free energy (which is like doing an extremely difficult integral by huge sampling), and that's of great interest for many people in physics and chemistry. (all of the papers you pointed out are published on JCP and PRL, all of them physics journals)

However, that is NOT the focus of the computational drug design. The focus of drug design is to predict WHAT drug binds on WHAT pose given a target structure so experimentalists can start making these compounds and do experiments. The exact binding affinity can be measured experimentally with ease. Providing the efficiency, I am not sure the molecular mechanics force field-based free energy calculation can be actually used in drug discovery someday.

---

Best UFO Resources
Wikipedia R@h
How-To: Join Distributed Computing projects that benefit humanity

...
or the THINK software prior to that.
...

Not just prior to that, but also after that with improvements
http://www.treweren.com/

(as a side note for anyone who participated in the Find-a-Druf project that use THINK as it virtual screening program. Some news (April 2006) if your interested http://www.treweren.com/news1a.html)

---

Team mauisun.org

-bump-

Per Rosetta's logo: Protein folding, design and docking

It'd be nice if we could learn more about Rosetta's use in docking and how it works in comparison to AutoDOCK, LigandFit, THINK etc (pls refer to 1st message).

---

Best UFO Resources
Wikipedia R@h
How-To: Join Distributed Computing projects that benefit humanity

-bump-

Per Rosetta's logo: Protein folding, design and docking

It'd be nice if we could learn more about Rosetta's use in docking and how it works in comparison to AutoDOCK, LigandFit, THINK etc (pls refer to 1st message).

Sorry about not answering this earlier. The improvement with protein-ligand docking in Rosetta is that
protein flexibility is modeled; this is not the case for most previous methods. The downside is that the docking calculations, while more accurate, are considerably more time consuming. The description of the interactions and the sampling protocol are very similar to the high resolution relax stage of the wu you are running now, except that small rigid body displacements of the ligand relative to the protein are also included

---