Dear Alden,
Thank you for your reply and your information. My idea is very similar to the idea presented in this nice paper. But the the approaches they used are beyond the limit of my knowledge. I hope some one can implement their results into a public available program (for example, Modeller, PyMol or Deep Viewer).
Direct homology modeling of phosphorylated polypeptide may not make any sense if the template used is not phosphorylated because the conformational change. But if a molecular modeling program allows to add the phosphate group on a model (similar to adding S-S bridge in Modeller) followed by MD simulation, the final model could make sense (Groban et al's paper is very encouraging).
I noticed that in the topological library of Modeller9v4, there are information for "converting (patching?) tyrosine to monoanionic phosphotyrosine (PRES TP1)", but no information on converting Thr, Ser and Asp. I wonder if any body tried to patch a model from Tyrosine to phosphotyrosine with Modeller. Is there any possibility to implement information for other residues for patching purpose?
Is there any other free program which has this functionality?
Thank you for your comments and suggestions.
Best regards,
Xiao-Ping
At 02:16 AM 10/29/2008, you wrote: >Perhaps homology modelling is not the best method. This might help: > >http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1440919 > >Best of luck. > >~alden > > >Xiao-Ping Zhang wrote: > > Hi, > > > > Many proteins are phosphorylated in vivo and and the status of their > > phosphorylation correlate to their function. The residues which are > > phosphorylated (Thr, Tyr or Ser) can be determined by mass spectrometry. > > I wonder how to build a phosphorylated model directly or how to apply a > > patch on a model to get a phosphorylated model in Modeller? > > > > Thank you advance for your suggestions. > > > > Xiao-Ping Zhang > > > > > > > > _______________________________________________ > > modeller_usage mailing list > > modeller_usage@salilab.org > > https://salilab.org/mailman/listinfo/modeller_usage > > > >
Respected sir,
I modelled number of proteins by basic modelling. After basic modelling during procheck evaluation the residues in most favorable region is generally above 90% but after loop refinement the residues in most favorable region comes below 90%. Please kindly inform me what may be the region for this.
Also after energy minimization the the percentage of residue in most favorable region decreses.
I will be grateful to listen from u regarding my query
Thanking you,
Dibyabhaba Pradhan Project Assistant, SVIMS BIC, India
On 10/31/08, Xiao-Ping Zhang xpzhang@ucdavis.edu wrote: > Dear Alden, > > Thank you for your reply and your information. My idea is very > similar to the idea presented in this nice paper. But the the > approaches they used are beyond the limit of my knowledge. I hope > some one can implement their results into a public available program > (for example, Modeller, PyMol or Deep Viewer). > > Direct homology modeling of phosphorylated polypeptide may not make > any sense if the template used is not phosphorylated because the > conformational change. But if a molecular modeling program allows to > add the phosphate group on a model (similar to adding S-S bridge in > Modeller) followed by MD simulation, the final model could make sense > (Groban et al's paper is very encouraging). > > I noticed that in the topological library of Modeller9v4, there are > information for "converting (patching?) tyrosine to monoanionic > phosphotyrosine (PRES TP1)", but no information on converting Thr, > Ser and Asp. I wonder if any body tried to patch a model from > Tyrosine to phosphotyrosine with Modeller. Is there any possibility > to implement information for other residues for patching purpose? > > Is there any other free program which has this functionality? > > Thank you for your comments and suggestions. > > Best regards, > > Xiao-Ping > > > At 02:16 AM 10/29/2008, you wrote: > >Perhaps homology modelling is not the best method. This might help: > > > >http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1440919 > > > >Best of luck. > > > >~alden > > > > > >Xiao-Ping Zhang wrote: > > > Hi, > > > > > > Many proteins are phosphorylated in vivo and and the status of their > > > phosphorylation correlate to their function. The residues which are > > > phosphorylated (Thr, Tyr or Ser) can be determined by mass spectrometry. > > > I wonder how to build a phosphorylated model directly or how to apply a > > > patch on a model to get a phosphorylated model in Modeller? > > > > > > Thank you advance for your suggestions. > > > > > > Xiao-Ping Zhang > > > > > > > > > > > > _______________________________________________ > > > modeller_usage mailing list > > > modeller_usage@salilab.org > > > https://salilab.org/mailman/listinfo/modeller_usage > > > > > > > > _______________________________________________ > modeller_usage mailing list > modeller_usage@salilab.org > https://salilab.org/mailman/listinfo/modeller_usage >
Dibyabhaba Pradhan wrote: > I modelled number of proteins by basic modelling. After basic > modelling during procheck evaluation the residues in most favorable > region is generally above 90% but after loop refinement the residues > in most favorable region comes below 90%. Please kindly inform me what > may be the region for this. > > Also after energy minimization the the percentage of residue in most > favorable region decreses.
I'm not sure what this has to do with phosphorylation, but there is certainly no guarantee that a single minimization will give a "better" model. You'd be better off building several models to get a reasonable sampling of the energy landscape and then picking the best cluster or single model.
Ben Webb, Modeller Caretaker
-
Dibyabhaba Pradhan -
Modeller Caretaker -
Xiao-Ping Zhang