I noticed that some loop modeling options (e.g. kic_looop_sampling) are now included for the enzyme_design.linuxgccrelease protocol in Rosetta3.3.
It seems that the capacity to remodel loops in the substrate-binding pocket will significantly improve the enzyme_design protocol. But since this has not been mentioned in any papers or tutorials, are the loop modeling options fully working now, or are they still under development?
Thanks for reading...
Florian would be the one to give the official response on the matter, but my understanding is that loop remodeling through the enzyme_design application is still in the development stages, and isn't "officially" released yet.
Note, however, that both enzyme design and some loop remodeling code is available through the RosettaScripts interface (See http://www.rosettacommons.org/manuals/archive/rosetta3.3_user_guide/Rose...), so if you are interested you can put together a protocol from the documented interfaces there.
there is some enzdes loop modelling stuff in the works, and the code is functional, however not documented yet, and no good real life experimental results have been obtained yet, but we're working on it:) the code is mostly in src/devel/enzdes/EnzdesRemodelMover. do you have access to that? i forget whether src/devel/ is part of the releases.
also, what exactly is it that you want to do?
To save the lzx32 the trouble, devel is not released.
Hey, many thanks for the replies.
Looks like I have to either wait for the new releases or bite the bullet and force myself to learn RosettaScript, which looks a bit intimidating for someoone with very little programming experience..
Hi Florian, we are hoping to alter the substrate-binding pocket of an enzyme to accept larger substrates by directed evlution (...and hopefully a little help from computational design). The crystal structure of the enzyme-substrate complex has already been determined. it seems that we won't be able to achieve the goal by simply redesigning and repacking the side-chains without changing the conformation of a few loops and even shortening one of the helices. I have been playing with with the enzdes protocol and reading the tutorials and papers from the Baker lab. But it seems it won't be very helpful at this moment without the loop modeling and flexible backbone functionalities.
since devel is not released, you also won't have access to the functionality through rosetta scripts. I could try to move the code into the next release version, but I don't know when that is going to happen, probably a few months from now.
what's your time horizon? do you want to get started on this right now or can you wait a few months?
also, as a precaution, my gut feeling is that for aggressive backbone redesign projects to work out, you'll probably have to be able to test 100s-1000s of designs. you mention you want to do directed evolution, so I presume you have a high throughput assay, but you also need to have a way of creating specific libraries. i.e. rosetta will spit out a bunch of models and then you need to synthesize the gene for each model. are you setup to do both the assay and the dna library synthesis in a high throughput way?
Hi Florian, it would be great if the new functionalities will be available in the new releases. I'll just wait then if there are no other choices.
We already have a HTP screening method in place for mutants screening. I thought about the size and cost of the specific DNA/gene libary you mentioned. Maybe i'm naive here, but my hope is that if I only touch the first and second-shell residues in the active site, the designs may converge and the top ones may share most of the residues. if that turns out to be true,we only need to have a few designed genes synthesized and the rest can be derived by mutagenesis...