changing the amino acid sequence within a project

[Andrea Piserchio]

Dear all,
a while ago I started a Project which now contains multiple spectra, a runabout assignment, and several lists manually assigned using "peak inspector".
After a while we realized that the sequence used to start the project was not correct.
The original one started as
GSHMEA...
but it should have been
SHMGELEA....
(bold where the 2 sequences finally matches).
I tried to import the correct sequence keeping the original numbering importing a seq file as shown below:
ser -1
his  0
met  1
gly  2
glu  3
leu  4
...

but the software seems nonetheless to re-number (not sure how much consistently) several peaks (but not all) within a list  (i.e, in a hhn noesy, [119.h, 120.h, 119.n] becomes [117.h, 118.h,117.n]; [160.he1, 160.hd1, 160.ne1] does not change).
Using the "clear PPMs and Atom Links" within the assign panel, and then reloading the ppm file does not seem to help.
Oddly this seems to happen after a while (after I open it with "show peak table"?),as the lists saved using "write all lists" immediately after uploading the new sequence are correct.
Is there a way out of this mess???

Thank you in advance for your help,

Andrea

[brucej]

Hi Andrea,

One way to fix this with minimal changes to existing data is the following (requires some manual editing).

Save the project so you can go back to a previous version if you mess things up.

Go into the projects star folder and edit the most recent .str file.

There are two parts you need to edit.  First, there is a single letter version of the sequence, like this (for ubiquitin):

_Entity.Polymer_seq_one_letter_code_can  ?

_Entity.Polymer_seq_one_letter_code

;

SHMGELVKTLTGKTITLEVEPSDTIENVKAKIQDKEGIPPDQ

QRLIFAGKQLEDGRTLSDYNIQKESTLHLVLRLRGG

;

Change that to be correct for your updated sequence (I modified the above ubiquitin sequence to have your changes as a test).

Then change the area that lists the sequence with three-letter code:

_Entity.Number_of_monomers            76

_Entity.Nstd_monomer                  no

_Entity.Nomenclature                  IUPAC

_Entity.Capped                  yes

loop_

_Entity_comp_index.ID

_Entity_comp_index.Auth_seq_ID

_Entity_comp_index.Comp_ID

_Entity_comp_index.Comp_label

_Entity_comp_index.Entity_ID

1 -2 SER . 1

1 -1 HIS . 1

1 1 MET . 1

2 2 GLY . 1

3 3 GLU . 1

4 4 LEU . 1

5 5 VAL . 1

6 6 LYS . 1

7 7 THR . 1

I modified the above to have the changes consistent with the sequence.  Note, I did not use a residue 0.  This can cause problems, so use a residue numbered -1 before residue 1.

The numbers in the first column are indices, and aren't used for anything.

If you have assignments that refer to the first few residues, and the new atoms are inconsistent with the names originally used, you may get errors when loading project.  So you may need to delete those entries.

Let me know if this helps (and feel free to pop over to my office and we can look at it together)

Bruce

[Andrea Piserchio]

Thank you,
unfortunately it seems that problem persists.
The peaks that are giving me problems have been probably labelled using the "Assign" button on the peakId window, so it would seem that somehow these are connected to the index.ID rather then to the index.Auth_seq_ID. Peaks get renumbered once I save the project and I re-open it.
For now I started a new project where I uploaded manually spectra, peak lists and chemical shifts as I need to finish up something quickly, but of course links infos are lost.
I can step by your office later this week if you have time.
Thank you again for your help,

Andrea

[brucej]

Yes, stop by my office when you get a chance.  I'm curious to see why this isn't working.  I did a quick test on a ubiquitin project where I changed the sequence in the same way yours needed changing and it did seem to work.

Bruce

[Andrea Piserchio]

My bad,
I did not realize you did not really renumbered the 1st column; if I do the same things seem to work now.
Should I change
_Entity.Number_of_monomers            165
to
_Entity.Number_of_monomers            167
?
Thank you again,
Andrea