You are here

Filter ddg_filter reports failure!

3 posts / 0 new
Last post
Filter ddg_filter reports failure!

I am trying to design a higher affinity interface for my protein-protein heterodimeric complex.But I'm getting the error which says "ddg is 0.246217 failing".I'm using the script similar to minpac_optE_premin.xml in the interface_design_demo folder of rosetta_2015.05.57676.

I really need some help to make this work.
Thank you!

p.s- I'll provide any other information if you need

Post Situation: 
Fri, 2015-02-27 18:19

The ddg filter in that xml file has a threshold of -15. That is, all poses that don't have a ddg -15 or better are thrown away. If you want to retain more poses--at least to make sure that your protocol is doing what you think it does--temporarily change that filter to a higher value (greater than e.g. 0.246217). Then you can lower it as you learn more about your system. (Maybe the wild type ddg is just -5 or something, so great redesigns might merely be -10!)

Also, how many structures are you making? The -nstruct flag controls how many, and since Rosetta's algorithms are Monte Carlo trajectories, some of them will always pass or fail some given filter.

Sat, 2015-02-28 13:50

Thanks for your reply!
1)I was making 50 structure and all of them resulted in the above error. The ddg for them was between -0.3 to -0.7 (I don't know why it is so low). But now I have changed the threshold value of ddg to -0.7 and am running the simulation with -nstruct 100 .

2)I have another question about the scoring function in this script.Is there any difference between using the 'score12' score function versus 'optE_inf_premin' weights that this particular script uses?

3)I have also used the online webserver RosettaDesign and it gave the following output file:
Mutation ddG_bind ddG_partnerA ddG_partnerB #Neighbors ddG_h_bond
187B L>W: -1.8 0.0 1.0 15 0.0
3A V>F: -0.9 -0.1 0.0 11 0.0
3A V>Y: -0.9 -0.1 0.0 11 0.0
3A V>W: -0.6 0.2 0.0 11 0.0
117B R>M: -0.5 0.0 -1.9 16 0.0
Do you know why it might be giving only these many mutations when the interface of the complex has over 20 residues?

Sat, 2015-02-28 19:24