[Demelerlab] updates to the UltraScan SOMO residue definitions

[Jason Zhu]

Hi Mattia,

Thank you for pointing out those formatting issues with the PDB! I tried
incorporating all of your suggestions, however I'm still getting errors
with non-coded atoms and missing atoms. I suspect the bigger issue is the
"ANISOU" label on a number of the atoms and how that affects the numbering
of the atoms. The "ANISOU" labeled atoms have the same number as the
preceding atom. Manually relabeling them isn't feasible as all of the
"CONECT" numbers would change accordingly.

I will talk to Adam in our lab who solved the crystal structure and see if
he can re-export the PDB file from Phenix refine so that it is more
compatible with SOMO. Let me get back to you once I have more info.

Best,
Jason

On Mon, Jul 31, 2023 at 12:35 PM Mattia Rocco <mattia.rocco at quipo.it> wrote:

> Hi Jason,
>
> one thing I've noticed is that in, for instance, the trimer x-ray file,
> ornithine is named "ORT" instead of "ORN". Plus, the ORT residues have "NA"
> instead of "N" as the N-terminal atom, and "CQ" instead of "CA". Also, the
> "XXX" (now MCA) has a totally different atom naming. Try replacing that
> naming with the one I coded in the somo.residue file, "HETERO" with "ATOM",
> and the it should work
>
> Best - Mattia
>
> Il 2023-07-31 20:56 Jason Zhu ha scritto:
>
> Hi Borries, Mattia, and Emre,
>
> I've attached the PDB files for the nmr models of the monomer, trimer, and
> dodecamer that should be correctly labeled and compatible with SOMO. I've
> also attached the updated residue and atom files courtesy of Mattia and
> Emre. I'm still having some difficulties getting US-SOMO to work on the
> xray pdb files due to how those PDB files are formatted and will get back
> to you as soon as I figure out how to make them compatible. Please feel
> free to reach out if you have any questions!
>
> Best,
> Jason
>
> On Thu, Jul 27, 2023 at 12:31 PM Mattia Rocco <mattia.rocco at quipo.it>
> wrote:
>
> Hi Jason,
>
> thanks, that's makes things much easier... ;-) :-)
>
> So, for Phi 5 / N-Me-Cha, in the PDB there are residues named ALC or CHA,
> without the N-Methylation. We could propose MCA, apparently not used for a
> residue [MHA is used, stands for
> (CARBAMOYLMETHYL-CARBOXYMETHYL-AMINO)-ACETIC ACID]. What do you think?
>
> Just an additional check for the psv: have you tried to reproduce a couple
> of the standard amino acid residues, just to be sure? Having done so
> countless times, I'm well aware of the potential pitfalls... :-( ;-) You
> can pick the values to be compared from the somo.residue file in US-SOMO,
> and the T-conversion formula is psv at 20C=psv at 25C+[4.25E-04 * (20-25)]
>
> Best - Mattia
>
> Il 2023-07-27 20:55 Jason Zhu ha scritto:
>
> Hi Mattia,
>
> Thanks for the questions! Here are my responses.
>
> 1. I did use the values from the 1994 paper that you listed. The values
> are for 25°C so they are not corrected for 20°C.
>
> 2. Phi 5 in the proposed NMR model is N-Me Cha. XXX in the x-ray derived
> model is also N-Me Cha. Would Xaa work as an acceptable 3 letter
> abbreviation for N-Me Cha? I am happy to correct the residue naming
> accordingly for both the NMR and x-ray pdbs if Xaa works!
>
> 3. I do have glycine where it is labeled as SAR and have corrected the SAR
> accordingly to GLY (attached).
>
> Best,
> Jason
>
> On Thu, Jul 27, 2023 at 9:37 AM Mattia Rocco <mattia.rocco at quipo.it>
> wrote:
>
> Hello Jason,
>
> I have started the coding for your residues, and I have a few questions
> for you.
>
> 1-In using the Durchschlag's values (I assume the Prog. Coll. Polym. Sci.
> 1994 paper), did you correct them for 20 °C? Because most are for 25 °C. I
> just need to know it, because all values in US-SOMO are at 20 °C.
>
> 2-I had no problems for Ornithine, which is a shorter version of Lysine...
> ;-) But I would need the real name for "Phi 5", and I don't see any N-Me
> Cha in the PDBs that Bo sent me. Is that the "XXX" residue? In that case,
> we need to pick a better three-characters name for it... ;-)
>
> 3-In those PDBs, there is "SAR", so I believe a proper psv should be
> computed also for this residue. If you have glycine, then the "SAR" residue
> should be changed to "GLY". Since US-SOMO somo.residue files are then
> distributed to the community, they should contain properly coded
> residues... ;-)
>
> Best - Mattia
>
>
> Il 2023-07-10 20:21 Jason Zhu ha scritto:
>
> Hi Mattia, Emre, and Borries,
>
> You should be able to use the following values for the non-canonical amino
> acids in the meantime. I assumed (-H2O) in determining the MW and partial
> specific volumes for the 2 non-canonical amino acids based on Helmut
> Durchschlang's papers.
>
> Ornithine:
> MW (g/mol): 114.2
> Partial Specific Volume (mL/g): 0.7795446
>
> N-Methyl cyclohexylalanine (N-Me Cha):
> MW (g/mol): 167.3
> Partial Specific Volume (mL/g): 0.90342716
>
> I would use the N-Me Cha values for Phi 5. For SAR at residue 13, I would
> use the MW and PSV for glycine instead. The analog that I based my NMR PDB
> model off of (PDB ID: 5v63) had sarcosine (N-methylglycine) where I have
> glycine.
>
> Hope that helps and please let me know if there's anything else you need!
> Jason
>
>
> On Mon, Jul 10, 2023 at 9:37 AM Mattia Rocco <mattia.rocco at quipo.it>
> wrote:
>
> Hello Bo,
>
> nice to read from you! I am currently away, I will be back in Genova a
> week from today, and I'll have a look at this residue coding task. I do not
> expect it to be difficult, especially if calculations of the psv of the
> non-coded residues are already available. In the meantime, you and your
> collaborators can use the approximate method, just enter the correct global
> psv and MW in the appropriate fields (Emre can point them out to you).
>
> Take care - Mattia
>
>
> Il 2023-07-09 23:52 Borries Demeler ha scritto:
>
> Hi Mattia and Emre:
> We are working with the Nowick group from UCalifornia, Irvine, to study
> synthetic amyloid peptide mimics which include several non-canonical amino
> acids. Loading Xray and NMR generated PDBs into SOMO causes several errors
> due to these discrepancies:
>
>
> Encountered the following warnings with your PDB structure:
>
> Chain B Molecule 1 Residue ORN 1: Non-coded residue.
>
> Chain B Molecule 1 Residue PHI 5: Non-coded residue.
>
> Chain B Molecule 1 Residue ORN 9: Non-coded residue.
>
> Chain B Molecule 1 Residue SAR 13: Non-coded residue.
>
>
>
> ...and various downstream issues with chain breaks resulting from these
> unknown residues. I suspect that due to the small size of these peptides we
> want to be as accurate as possible to represent the residues correctly,
> since they are likely extending into the solvent interface. Of great
> interest in amyloid research is the structure & function of small soluble
> oligomers, so this is what we are trying to investigate by AUC, NMR and
> X-ray, and being able to simulate expected oligomers by US-SOMO would be a
> great plus.
>
>
>
> Can you please update the residue definition list for US-SOMO so these AAs
> are properly recognized? I think this would be a great and novel use of
> US-SOMO!
>
>
>
> I have attached a number of PDB files from oligomeric structures of one of
> these peptides derived from NMR and Xray that were generated by Jason Zhu
> (Cc'ed) from James Nowick's lab, and since I am no expert in PDB formats,
> perhaps you and Jason can collaborate to get the residues added, and allow
> us to load them correctly?
>
>
>
> We would be happy to include you in the manuscript author list for your
> help with the simulation. The s, D and frictional ratios would be directly
> compared with AUC data. Also, James has already worked out the partial
> specific volumes for these non-canonical amino acids using Helmut
> Durchschlag's papers. He can provide you with these values.
>
>
>
> Thanks so much, -Borries
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.biophysics.uleth.ca/pipermail/demelerlab/attachments/20230731/776ac86a/attachment-0001.html>