
   For Release 0.8.9.3

   o Refine a bit of distorted RNA
     Marc Leibundgut 26/06/2019.

   o how about fixing the raster3d install? Make it dependent on having gfortran

   o Fix the representation of charges in the residue restraints notebook

   o multiple file selection in the dictionary cif chooser (don't spend too much time on this)

   o Fix the H5' problem

* Coot works - the dictionary is wrong.

   o add a function to set the window title (don't for get to check
     for use-graphics)

* set_main_window_title() already exists.

   o what is the map grid? Tell me in the map preferences.

* Done.

   o fix the compilation of libcurl in the build script to use
     libidn2 and OpenSSL for https

* Done.

   o Rob's JSON file for mtz animation - do we have the JSON parser here?

* Yes. But you need to test for it in the same way that it's tested to set BUILD_CURLEW
  Give up with this for now - too messy because the file names change.

   o Refinement on an alt conf should refine just the atoms with that alt conf.

   o for drawing bonds - don't make a S-S bond if the S is linked to something else.

* In refinement.

   o Huw (from York) sent a message about pygtk usage. Needs to be incorportated. ::: 1

   o Fix up dialog in RCrane, like above hopefully.  ::: 1

   o Read in a PDB file that has a TER. Write it out as a mmCIF. It has a blank atom at the
     end - it should not [Nigel Moriarty]. mmdb bug? ::: 1

* It is still there in Coot. It's most probably an MMDB bug. How to test this issue with
  a CCP4 program? I can't get coord_conv to work.  Mail Eugene.

   o other people running the crash catcher. Is it sufficiently robust? coot-experimental?
     Run the hal-built binary on pc4



   o use stroke characters and pull out the window. Can this be fixed?

* Yes. Done.

   o read pdb3siu.ent and write it out as a cif. ::: 2

* Fixed.

   o add a property/timeout to a map (or model) that looks at a file and if it gets updated
     then it recreates itself from that file.

* Done

   o Add a residue to a residue that has a TER. The TER is kept on the added-to-residue.
     It should not be. [Christine Gee] ::: 1

* Fixed already.

   o Refine a bit of distorted RNA: Marc Leibundgut 26/06/2019.

* Won't fix in 0.8.9.x.

   o crash: Use CCP4 Coot, Accept Baton Atom then JED flip. ::: 0.5

* Can't reproduce.

   o Patterson for Randy

* Done.

   o Make a "I like the graphics size and position" - like for dialogs.

* Done.

   o if balls and sticks representation then draw balls for waters, not sticks

* Done.

   o Make a "I like the graphics size and postion" - like for dialogs.

* Done.

   o backup format change to mmdb-binary.

* It is now cif if input was cif, that will do for now.

   o Find Ligands: if the ligand was not specified, don't dismiss the dialog (unless there are
     no ligands)

* Done.

   o density fit graph should say both the map and the model in the title

* Done.

   o Refine residues with alt confs (e.g. A71)

* Done.

   o Message from Charles:

Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0   libcoot-geometry.0.dylib 0x0000000101816419 coot::operator<<(std::ostream&, coot::residue_spec_t const&) + 9

* Done

   o Torsion general - message if atoms are not in the same residue

   o Torsion general - message if atoms were double-picked

   o validation outliers gui should order by residue number, box badness indicator?

   o validation outliers gui "don't show me this again" button/mechanism

   o Fetch molecule favipiravir doesn't work

* It does now.

   o fix the colours of (say) 5vic in Calpha Backbone mode

* Done.

   o refine residue ranges with ins codes (antibodies)

* Seriously consider not doing this - it is terrible spaghetti - and works in the 0.9 rewrite.

   o Make the ligand fitting progress bar work for the fitting part (use 2 progress bars?)

* Punt. Nice to have, but too much rewiring. Sorry Keith.

----

   For Release 0.8.9.2

   o fix clipper configure fftw2 - Don't use the system version.

   o Can you do "x" refine on a residue with a insertion code, ee.g. H 35A in 5aco.  Triple Refine this too.

   o glyco tree refine A 1334 in 5aco.

   o fix the colours of (say) 5vic in Calpha Backbone mode

   o make anamalous maps non-red/green

   o edit chi angles gives the wrong chi angles for chi2? update_residue_by_chi_change() uses
     tree.rotate_about() with 2 atoms as the arguments.  tree should preferentially
     choose N as the base atom, if it is an option. Also, CG1 rather than CG2.

   o validation outliers gui in the release, in the gui.

   o Fetch molecule favipiravir doesn't work

   o fix the colours of (say) 5vic in Calpha Backbone mode

   o Dock sequence?

* Hmm...

   o read in a phs file - can you scroll the map? No, but + and - work.

* Done.

   o add --command in --help

* Done.

   o make the screen output of peak-search give the Nsigma also.  BR

* Done.

   o All symmetry to always be on (add a setting to ~/.coot-preferences)

* Done.

   o Difference map Peak search results came from which map? Put it in the titlebar
     EJD

* Done.

   o make anamalous maps non-red/green

 * Done.

   o twisted peptide/geometry omega added to validation outlier dialog.
     does cis_peptides(imol) do what is needed?

* Fixed.

   o can you make a stand-alone doxygen version of the python documentation?

* In part - python files need to be added.

   o Does align and mutate work?

* It does now.

   o multisharpen in a directory not writable. Give a better error message

* Done.

------


   o delete is_sane_inter_residue_distance debug

* Done.

   o Copy Fragment (new_molecule_by_atom_selection()) should copy links too.

* Done.

   o Andreas K:

   > Coot now adds new ligands to the chain of the corresponding protein. Refmac5 (5.8.0238) then adds two TER
   > records to that chain, one after the protein and one after the ligand. When now adding a C-terminal residue
   > in Coot, it will not remove the original TER record after the protein

* Can't reproduce.

   o something for the trip: The acedrg link dialog needs to be able to delete a bond
     and change a bond order
     Needs to be done in python too

* Done for the moment, needs Bernie to do python version.

   o Phosphorylate this residue on a SER. New dictionary problems? Problem is in refinement part. OXT?
     What happens if there is no map? We should get a regularized model.

* Done.

   o Add residue type to the validation outlier ramachandran outlier button
     (should the rama plot and the rama outlier have the same criterion?)
     Needs to be done in python too

* Done - needs python update.

   o loops: when the gap between residue number is small but the gap is large, we
     need to be able to represent that differently. With big red discs - or something.
     John Berrisford.

* Done.

   o OC: Using delelete residue range for all residues in the moleucle should delete the
     molecule from the display manager.

* Done

   o fill and use go_to_ligand_non_interesting_comp_ids: add_non_interesting_ligand_type().

* Punt

   o Map Dialog, Setting 8 rmsd should move the block in the graph below.

* Punt

   o check for double ligands and/or nan atoms when using set_merge_molecules_ligand_spec().
     Add it to the test (do I have a test already?)
     Check that the returned value corresponds to the new ligand: L 1

* Done.

   o Try rotate/translate on a chain by picking C4',B in B 2 of 2H0N

   o update the copyright in the about dialog?

   o Randy: update NCS matrices for "O" key after symm shift reference chain here

* fill_ghost_info() needs to be called again to regenerate the matrices.
  Done.

   o this was from coot-experimental but maybe it's a problem here.

#7  0x00002b2579376601 in char* std::string::_S_construct<char const*>(char const*, char const*, std::allocator<char> const&, std::forward_iterator_tag) () at /lib64/libstdc++.so.6
#8  0x00002b25793769b8 in std::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(char const*, std::allocator<char> const&) () at /lib64/libstdc++.so.6
#9  0x00000000007a0b91 in graphics_info_t::save_directory_for_saving_from_filechooser(_GtkWidget const*) (this=this@entry=0x7ffcc09e7d3f, fileselection=fileselection@entry=0x16c474c0) at graphics-info-gui.cc:665
#10 0x000000000069f1ab in save_directory_for_saving_from_filechooser(GtkWidget const*) (fileselection=fileselection@entry=0x16c474c0) at c-interface.cc:1466
#11 0x00000000006cd628 in on_save_coords_filechooserdialog1_response (dialog=<optimized out>, response_id=<optimized out>, user_data=<optimized out>) at callbacks.c:10897

* Done as much as I can from this, I think

   o somewhow draw the baton so that the direction is obvious. As an arrow, perhaps?

* Done

   o Delete extra restraints - does that work?

* Yes - I wonder what I thought the problem was.

   o Capture the mmdb error message ("Disk quota exceeded or some such") when
     trying to save/quick-save a pdb file).

* There's already a dialog on error on write - that will do for now (it doesn't say the error type)

   o when multi-sharpen fails because there is no refmac, we should get a message
     in the status bar.

* Done - something like that.

   o Add map resolution from grid spacing to Map Properties dialog - Bernhard Rupp.

* Done.

   o when generating density lines, check for is-EM map, and if so, don't draw outside
     the boundaries of the cell.
     This needs to know that a reboxed map is *not* an EM map because it can have density
     outside of the unit cell - to include reboxed map, it needs to have x,y,z limits
     derived from the grid box.
     Remove the current box restriction and move to refinement

* Done.

   o zero occupancy markers should respect the model molecule display limit

* Done.

   o zero occupancies markers for hydrogens should respect the no-bonds-to-hydrogens
     setting

* Punt. Too hard too esoteric.

   o accept_regularizement() needs a better name (keep this one for backwards compatibility)

* Done.

   o Label for "Limit Model Display Radius" should have a ...

* Done.

   o N-linked Additon -> remove the word "Dialog"

* Done.

   o Oliver-Clarke bonding

*Done! Yay.

   o Pointer distances should not show distances to Hydrogen atom if they are not displayed!
     Steven Sherrif

* Done.

   o do the bug fix for Robbie.

* Done.

   o CA-Zone -> mainchain on a zone that includes negative residue numbers -> crash
     Oliver Clarke. ::: 2 h

* Done.

   o remove default adding of monomer on read cif dictionary

* Done.

   o when I have undisplayed hydrogens, I don't want to ever pick them during refinement.
     Fix. ::: 1

   o G is reverse JED
 
* Done.

   o While we are at it, I don't want to pick atoms at the back when I am refining either.
     ::: 1
   o Fix db_mainchain fragment/chain indexing bug. Oliver Clarke ::: 1 h

   o copy fragment should have a "move it here" button. ::: 0.5

   o Draw -> Cell and Symmetry -> symmetry colour should follow the
     real symmetry colour. But now the symmetry colour is not dynamic.
     Swings and roundabouts.

     ::: 1

   o make a patch for clipper hkl_data.h line 421 :: 0.5

   o graphics_info_t::read_standard_residues() uses
     std::string standard_file_name = PKGDATADIR;
     it should use
     std::string l = package_data_dir();

     ::: 0.5

   o Make Fe-S cluster SF4 (and FES?) drawn with bonds ::: 1

   o Fe-S needs to be rendered nicely in Bonds mode ::: 2

   o Add anom to PDB-redo output ::: 0.5

   o dictionary_radii() for Hamish. Return something JSON.
     a dictionary? with a dictionary of atoms the the radius as the data :: 1

   o Coot found no unexplained blobs" dialog should have a better title
     than "dialog1" ::: 0.5

* Done, needs testing.

   o crash on coot-reduce on luigi/3p4gA_dom1.pdb ::: 1

   o Kill off Lidia -> View in Lidia, prodrg path ::: 0.5

   o this is running coot-refinement and pressing Apply in Lidia

INFO:: backup file coot-backup/0A_from_dict_Thu_Apr_12_15:33:04_2018_modification_0.pdb.gz
[15:33:04] Explicit valence for atom # 21 O, 3, is greater than permitted
std::exception
There are 2 data in /Users/pemsley/autobuild/build-refinement-pre-release-gtk2-python/share/coot/lib/data/monomers/a/A.cif
sys:1: GtkWarning: Unable to find default local directory monitor type
command "cprodrg" not found

    ::: 1

* Bleugh. Bye bye.

   o Jiggle fit uses linear interp? Jiggle fit should select backbone
     if fitting protein, ring atoms for RNA/DNA ::: 2

* Done. More tricky than expected. Nice.

   o make restraints for a big protein, then delete them using ...
     delete deviants(?) Fix the slowness ::: 2

* Done. Nice.

   o morph fit by secondary following the cryo-tutorial, morph secondary
     fit makes a helix go wonky. Check with self chain restraints [JED]. ::: 0

* Can't reproduce.

   o reposition the O on the current residue when adding a new one -
     Add Terminal Residue execute_add_terminal_residue() ::: 1

* Done.

   o greg-tests: "Correction of CISPEP test"
     Don't run grep ::: 0.5

* Done. call-with-input-file read-line port

   o Add LINK info to header browser. Other things? ::: 2

* SSBOND bonds are not part of mmdb, BIOMT is a REMARK 350 and is already there.

   o seems to be a failure of peptide plane restraints when residue type is a PRO.
     (converted to a ALA and things flattened out as expected). But is this a problem
     only in refinement branch? ::: 1

* Add planar peptide restraints for PTRANS. Yay.

   o maybe add external angle restraints. Make them work in mini-rsr too ::: 2

* Done. Nice (tested).

   o Eleanor's HIS. Add test. This is tricky, it uses mgtree ::: 2

* Done with a bit of a hack in atom ordering.

   o Make RCrane work again ::: 2

* Done - residue_info() had changed.

   o Update the density fit graph on delete zone ::: 1

* Done.

   o fix crash in all_molecule_ramachandran_score_py() ::: 0.5

* Done.

   o put this in the status-bar:

     Sorry atom with name " CA " alt-loc "empty", res-no: 20, ins-code "", chain: "A" not found in molecule 0

    ::: 0.5

* Done.

   o replace fragment using Edit gui menu item ::: 0.2

* Done.

   o sync bernie ::: 1

* Done.

   o add labels to raster3d output ::: 2 h

* Done.

   o after a merge molecules, do a graphics draw ::: 0

* Don't see the problem

   o fix this in molecule-class-info.h:
      // Ugh.  Horrible.  I don't want outside access to setting of
      // imol_no - I want to do it in the constructor.  Must FIX.
      //
      void set_mol_number(int i) ...
      ::: 1

   o Why no ligands in "CA + Ligands" on the Slt pdb? ::: 0.1 h

* Because the ligands don't match the dictionary. Is fine.

   o Does replace residue get the residue numbering right?
     Change the ALA at the end of a chain ::: 2

* Done. A solid 2. Took all day. 4.46pm now.

   o test that add-lsq-atom-pair and apply-lsq-matches works ::: 0.2

* Yeah it does - "LSQ by atom" is the test.

   o links (of renumbered residues) disappear on renumber residues ::: 1

* Done - update_any_link_containing_residue()

   o      ERROR:: C: last_restraints no cleared up
     on every frame of a rotate/translate molecule ::: 1 h

* Done

   o coot in phal gives old (0.8.8) coot. ::: 0.1 h

* Already Done? Now gives release version, as it should.

   o Add OXT to residue -> on a DAL -> triple bond? ::: 1 h

* It was because I was adding an ATOM to a residue with HETATM. Now the
  OXT is a HETATM too, if needed.

   o curlew on hal? ::: 1 h

* Seems OK.

   o email about dynamic bond width from Erec Stebbins. ::: 3 h

* Done. Nice. I will use it.

   o does RCrane have a menu item again? ::: 1 h

* Done.

   o peptide flags needs to obey the radius limit too - if it was set. ::: 1 h

* Done.

   o fix the glib bug:
     https://stackoverflow.com/questions/23199699/glib-critical-source-id-xxx-was-not-found-when-attempting-to-remove-it
     ::: 1

* Done?

   o Add coot-extended.

* Done.

   o Merge bernie  ::: 1 h

* Done - I hope

   o coot --version and --version-full ::: 1

* Works.

   o Authors and Journal in header browser? ::: 3 h

* Done.


   o SMILES -> Simple 3D ::: 1 h

* missing move_molecule_to_screen_centre_internal()

   o does multi-sharpen work?

* yes.

   o the application of a pir alignemnent needs to add a save.

* done.

   o mutate an ALA with hydrogens to a GLY. HA floats away (it should be deleted).
     Make a test.

* Done.

   o Use fragment/linked-residues behind the CA-Zone -> Mainchain button

* Done

   o Mutate DNA in Mutate Residue Range (DNA generate in Ideal DNA/RNA)

* Done.

   o use comma to split fields in search monomer library. Flight-time fun.

* Done.

   o cherry-pick e5278d7fad21a8813cccb78ed51b5cc038947a63 from refinement

* Done.


   o FLEV on A401 NPD of 5uat (after Hydrogenate region)

* What was the problem? Looks OK to me - apart from Star Wars.

   o distance, symmetry atom click first -> wrong first atom

* Done.

   o Don't use " CG " quoting in the cif file output by Edit->Restraints->Write

* Done.

   o sync Bernie

* Nothing new.

   o does the tree comprehension work? (Marilyn Yoder problem)
   o is the carbohydrate module in a working state?

*it does now.

   o ASN has Hydrogen atoms. How does cho builder work?

*it does now

   o 1abr 140B "Delete all carbohydrate" fail

* Insertion codes!

   o multiple maps - undefined refinement map - jiggle fit this molecule needs an dialog alert.

* Done.

   o When merging a ligand - allow the ligand to be in user-specified chain.
     Maybe a menu item "Merge this ligand" which uses new function
     merge_ligand(imol, current_residue_spec, into_mol, new_residue_spec)

* Done. Even added a test.

   o Make a trivial ligand in the ligand builder, run acedrg. make an incorrect ligand
     (valence error) and run acedrg - the previous ligand gets imported.

     -> Apply button should grey out if there is no SMILES string (perhaps)

* Done.

   o add-terminal-residue should do not make intermediate atoms and not do a rigid body fit.
     Use --no-startup-scripts to help test this.

* The second of those was already so, immediate addition mode changed.

   o Carmen question: ?

       If you are on the keyboard and doing rotamer, can you accept it with Enter?
       No.
       Fix it then.

* Done.

       Adding solvent molecules - add it to the right molecule - make the spacing bigger to make the widget bigger.

* Done.

   o Draw hydrogens is not respected in colour-by-chain mode.

* I wonder what I meant. It was respected, but colour-by-secondary-structure was not. I had been looking at the
  refinement build, which has (had?) different arguments. Fixed.

   o Add a test that fle_view with_png output makes a sensible file.

* Done

   o Add distance restraint - does that work? - What about helical restraints?
     Test with a distorted model in a very low resolution map.

* It's working fine. There needs to be a lot of them to make a serious impact on the refinement.

   o what has happened to RCrane?

* It has moved to Calculate

   o I don't like the double call to make_bonds() (it's in CA mode at least). Get rid of it.

* I don't see it.

   o OXT in residue type LYS fle_view_with_rditk() failure.
     New monomer library!?

* Punt.

   o Randy: origin removed patterson: what's next to fix?

Punt.

   o compare the origins of a rebox and maps from the EMDB - look for nxstart etc differences
     use mrcfile for convenience.

* Punt.

   o CryoEM tutorial says "choose a sensible weight for the map" - how do we do that Paul?

* Is this really a problem - if so, punt.

   o add symmetry and try povray - crash?

* Punt.

   o raster3d should show symmetry?

* Punt.

   o can you add a download for EMDB bundles?

* Punt.

   o does normal/old/standard coot cis-trans update the bonds?

* Yes
