about ppi design about proteinmpnn

[jadolfbr]

Did you try fast relax of the WT and the design? Were there any clashes? How did you do the threading?

…

On Wed, Aug 3, 2022 at 12:24 AM 18217265596 ***@***.***> wrote: hi proteinmpnn team i read the paper about the function design part i wonder if protein mpnn can be used to design ppi in certain interface i tried it with one side of the interface with residue within 5A of the other side into design and other fixed the designed sequences are predicted of their structure and docked to the other side but ddG calculated by interface anaylzer with beta nove 16 is positve but wt is much negative is there anything to discuss about such situation? — Reply to this email directly, view it on GitHub <#15>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAZDHRAASB6NSJ5R5OG5QFTVXHYBJANCNFSM55NLUB7A> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[18217265596]

Did you try fast relax of the WT and the design? Were there any clashes? How did you do the threading?
…
On Wed, Aug 3, 2022 at 12:24 AM 18217265596 @.> wrote: hi proteinmpnn team i read the paper about the function design part i wonder if protein mpnn can be used to design ppi in certain interface i tried it with one side of the interface with residue within 5A of the other side into design and other fixed the designed sequences are predicted of their structure and docked to the other side but ddG calculated by interface anaylzer with beta nove 16 is positve but wt is much negative is there anything to discuss about such situation? — Reply to this email directly, view it on GitHub <#15>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAZDHRAASB6NSJ5R5OG5QFTVXHYBJANCNFSM55NLUB7A . You are receiving this because you are subscribed to this thread.Message ID: @.>

I did not fast relax WT , but WT crystal structure was used in flexddg in designing interface residues and computational results were fine. no clashes in wt anyway
and sorry i dont understand what means "do the threading" can you explain it a little?
in proteinempnn i just used the scripts Example from vanilla_proteinmpnn/examples/ to design some complexes
and changed the residue number to my interface and the other side of the interface fixed

[18217265596]

Did you try fast relax of the WT and the design? Were there any clashes? How did you do the threading?
…
On Wed, Aug 3, 2022 at 12:24 AM 18217265596 @.> wrote: hi proteinmpnn team i read the paper about the function design part i wonder if protein mpnn can be used to design ppi in certain interface i tried it with one side of the interface with residue within 5A of the other side into design and other fixed the designed sequences are predicted of their structure and docked to the other side but ddG calculated by interface anaylzer with beta nove 16 is positve but wt is much negative is there anything to discuss about such situation? — Reply to this email directly, view it on GitHub <#15>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAZDHRAASB6NSJ5R5OG5QFTVXHYBJANCNFSM55NLUB7A . You are receiving this because you are subscribed to this thread.Message ID: @.>

also im not sure if im doing it the right way to use the proteinmpnn designed sequence to predict structure in colabfold and align to wt to calculate interface ddg, since proteinmpnn has no structure output or is such calculation just wast of time before experiment of the designed sequence?

[gezmi]

You will need to optimize (relax with BB and maybe with side-chain restraints) AF2-predicted structures, they are in a different force-field than X-ray structures, so Rosetta scores will always be very positive for them. Even better, just get the X-ray structure and thread the newly designed sequences onto them (you can do this with flexddG), and then score.