-
Notifications
You must be signed in to change notification settings - Fork 2
/
Changelog
executable file
·119 lines (113 loc) · 6.73 KB
/
Changelog
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
* Ligands such as MAP contain a metal atom (one of those handled by SLIDE and
SimSite3D). These metal atoms were considered to be part of the protein
which is not correct. This problem has been "fixed" in the sense that
such metal atoms are no longer considered as part of the binding site.
Whether this is more or less correct than before is debatable since
in some of the cases the crystallographers may have chosen to use such
ligands in the absence of or to displace bound metals required for
catalysis.
* Added a check to force His ND1 to only be DONOR in terms of salt bridge
checks.
* Changed His NE2 to be donor only -- as per Jack Kyte's Structure in Protein
Chemistry pKa diagram for the high-probability states of His around pH 7
* Fixed issues with HETATMs if HET records are NOT in PDB file. They are
complained about and then ignored for surface generation. Thus if you
have modified residues near the binding site and want them to contribute
to the molecular surface they MUST have MODRES entries in the protein
structure (PDB) file.
SimSite3D_Software_2.7.2
* Cleaned up misc. files there were no longer used.
* Adding support for metals in sitemaps
* Repaired issue with fragmenting mol2 files with multiple substructures.
* Atoms which were labeled as HETATMs in the input PDB files should now be
correctly labeled as HETATMs in any output PDB files.
* Having the names of runtime variables in the config files not prefixed with
SIMSITE3D_ was deemed a possible place of confusion. The example config files
and parsing was updated accordingly.
* Blank chainIDs and iCodes are ommitted (not represented by an character) from
the label in the sitemap labels file. If an iCode is not a blank, it is
appended to the residue number; if a chainID is blank we get "__" instead of
"_C_" where C is a wild card for a nonblank chainID.
* Added support for waters to be considered as part of the protein when
generating a sitemap.
* Mucked around with templates for the scoring functions.
* Template points now have a unique label in the corresponding csv file.
SimSite3D_Software_2.7.1-r1
* Fixed copy-n-paste bug in method used to load environment variables
* Programs no longer allow explicit loading of conf files from command line
* conf files are no longer required
* system conf file now is /etc/simsite3d/simsite3d.conf
* local conf file now is ${HOME}/.simsite3d/simsite3d.conf
SimSite3D_Software_2.7.1-A6
* mol2File::write() now puts the "struct_id" + _NNNNN as the molecule name
(instead of "*****"). mol2File::write() now places the atom-type + orbit
in a 6 char column.
* We cannot assume that the input files (structures, ligands, etc) will be
created only on *nix platforms. As such, the mol2File class had to be
modified to ignore the '\r' Windows includes to denote the end of a line.
* The environment is handled in a somewhat general fashion so that it is
relatively easy to add or change environment and configuration variables.
Of course, this has nothing to do with the actual implementation of the ideas
the variables represent.
* Renamed some of the SimSite3D Software cmdline parameters (flags) to better fit
the updated cmdline options
* Ligand fragments with few than N heavy atoms are discarded. If no ligand
fragments remain, the alignment is discarded.
* Looking at a radius of 1.5(A) around template points shows that (especially
for pockets) a number of small holes exist in the "interior" of the
template volume. Bumping up to 2.0(A) seems to fill in the holes and should
not increase the volume much. However, this method is realtively costly
and is not ideal. If we could use an ellipse or set of bounding boxes to
represent the template volume, it is likely we would be better off.
* Added conf file, environment variables and conf file on cmdline
* Fixed bug that was putting additional atoms in the sitemap atoms file
* Added environment variables (and the associated conf entries) for the
diverse dataset and ligands
* Sitemap .csv files only have the name of the files and not the path to the
ligand and protein files. In addition, we assume that all of the sitemap
files are in the same directory. Given that assumption we do not need to
write the _s.pdb and _a.pdb and _rad.pdb in the XXXXXX_s.csv file.
* Repaired the install.py script so that it does not install the
auto_gen_sitemaps.pl script, does not install the diverse sitemaps and
does not blow away the python and examples directories if it is installing
to the source dir.
* Changed the way that the external scoring functions are called. The explicit
command line is listed in external_scoring_functions.txt with $PROTEIN and
$LIGAND denoting where SimSite3D will stuff in the chosen protein and ligand
files.
SimSite3D_Software_2.7.1-A5
* Version of software is now added to the sitemaps .csv files and the search
results files and stdout.
* Diverse ligand that had a Br labeled at Du is fixed -- MolCharge (or its
forcefield) does not know how to handle Br atoms.
* Allow S.o2 and S.o as aliases for Tripos Mol2 atoms S.O2 and S.O -- why did
Tripos decide these should be capitalized?
SimSite3D_Software_2.7.1-A4
* Changed the first field of a score record (line in output flat file) to be
that of the ligand fragment (the ligand fragment file can be found in the
corresponding ligand_fragments directory -- typically the same directory
as the output file).
* The ligand fragments file names have been changed from XXXX_NNNNN_frag_l.mol2
to XXXX_NNNNN_f.mol2 (where XXXX is the structure id and NNNNN is a 5 digit
zero padded orientation/alignment/score number for the given structure id).
* The moved ligands file names have been changed from XXXX_NNNNN_SimSite3D_l.mol2
to XXXX_NNNNN_l.mol2.
* The check for the number of database hit heavy ligand atoms in the query
pocket now correctly disregards the number of hydrogen atoms in the pocket
* Added a fix to separate the atom and substruct id fields in mol2 if they are
smashed together.
* The gen_points and search_sitemaps files now support the command line options
'-V' and '--version' to print out the version information of the program.
SimSite3D_Software_2.7.1-A3
* Added error checking to give reasonable warning messages and exit when
required parameters are not listed to gen_points.
* mol2 ligand hydrogen atoms are now excluded when using mol2 ligands to
define a sitemap volume.
* Removed python dependencies on versions of python >=2.3. We should be
able to get by with using items from python >=2.2.
* Fixed some off by 1 errors for processing/checking comand line arguments
* Added somewhat reasonable checks for 3pt alignments which are numerically the
identity alignment.
SimSite3D_Software_2.7.1-A2
SimSite3D_Software_2.7.1-A
SimSite3D_Software_2.7.0