You are here

the flexpepdock refinement and ab-initio

7 posts / 0 new
Last post
the flexpepdock refinement and ab-initio
#1

Hello,
I applied the FlexPepDock refinement protocol and the FlexPepDock ab-initio protocol to simulate the same peptide-protein complexe(1K5N) and generated a mass of models(refinement n=1000; ab-initio n=50000). The score function of two protocols is score12. The models was ranked based on the reweighted-score.
However, the score of the top models of refinement is better than ab-initio.Sub-angstrom models were sampled in the cases of refinement. Even though ab-initio protocol generated near-native models,their energy score is high.
In ab-initio i don't observe the trend that near-native models have low energy.In most of cases the top models have unqualified rmsd.
The scatter diagram have been attached.

AttachmentSize
1ogt.pdf216.25 KB
1k5n.png583.81 KB
Post Situation: 
Mon, 2013-12-23 19:17
tfliu

Hi tfliu,
The results are quite unexpected given the same initial structure for both the runs ( same initial structure,right? ). Could you please provide the commands you used for running both the refinement and ab initio?
I am hoping you have mentioned chain C as the peptide chain ( by default it's always the second chain).

Tue, 2013-12-24 04:00
nawsad

Hi nawsad,
Refinement:
I place the mutant peptide in the receptor binding site. The bb-RMSD of staring structure is 3.1.
Flag
-s 1k5n.pdb
-native native.pdb
-pep_refine
-ex1
-ex2aro
-use_input_sc
-nstruct 1000

ab-initio:
I apply BuildPeptide.linuxgccrelease to generate extended peptide and place it in the correct binding site.The bb-RMSD is 4.6.
Flag
-s 1k5n.pdb
-native native.pdb
-receptor_chain A
-peptide_chain B
-nstruct 50000
-lowres_abinitio
-pep_refine
-ex1
-ex2aro
-use_input_sc
-flexpep_score_only
-peptide_anchor 285
-frag3 frags.3mers.offset
-frag5 frags.5mers.offset
-frag3_weight 1.0
-frag5_weight 0.25

Tue, 2013-12-24 18:33
tfliu

Hi Tfliu,
everything seems OK to me. Could you please upload the pdbs if possible.

Wed, 2013-12-25 01:25
nawsad

Thanks.
Here are my pdb files.

Thu, 2013-12-26 07:09
tfliu

Hi tiflu,
Seems like sampling is the issue here. The scoring function is working fine and refinement of native gives structures more stable that what found from the ab-initio docking. There are two probable reasons I see, one being in the creation of fragment files. As the predicted secondary structure is helical for few residues which is not in real case. May be the initial structure is trapped in a funnel which is quite stable and not able to get back to native with default perturbation values. Could you please rerun with increased perturbation values? Also you can run another run with different fragment files. I will say use fragment picker and use less priority and weight for secondary structure score.
here are the flags you can use to run with increased rigid body and BB sampling.

-random_phi_psi_preturbation 40.0
-random_trans_start 4.0
#-random_rot_start 10.0
-smove_angle_range 30

best,

Sat, 2014-01-18 07:44
nawsad

hi nawsad! Thank you very much!

Sat, 2014-02-15 03:56
tfliu