In Dragon 5.5 two new blocks of molecular descriptors were added, leading to a total of 1560 (780x2) new descriptors.
The new block 21 is comprised of 2D binary fingerprints, which define the presence (1) or absence (0) of specific atom pairs at different topological distances (from 1 to 10). The new block 22 is comprised of 2D frequency fingerprints, which differ from the binary fingerprints since the frequency of specific atom pairs at different topological distances (from 1 to 10) is calculated. The considered atom pairs are the combinations of C, N, O, S, P, F, Cl,
Br, I, B, Si and X (where X represents any other atom, except hydrogen). Changes between Dragon 5.3 and Dragon 5.4. Changes between Dragon 5.1 and Dragon 5.3. Changes between Dragon 3 and Dragon 5.1.Changes between dragonX 1.2 and dragonX 1.4
dragonX has been deeply improved in order to make the descriptor calculation faster. As an example, both dragonX 1.2 and 1.4 have been tested on 3276 molecules. These moelcules have been taken from the NCI Enhanced Database as molecules including the experimental logP. The test has been performed on AMD Athlon(tm) XP 1600+ (RAM 256 MB) (Ubuntu 7.10). dragonX version 1.2 required 9m and 45s, while dragonX version 1.4 required just 5m and 39s.
In dragonX 1.4 two new blocks of molecular descriptors have been added, leading to a total of 1560 (780x2) new descriptors. The new block 21 is comprised of 2D binary fingerprints, which define the presence (1) or absence (0) of specific atom pairs at different topological distances (from 1 to 10). The new block 22 is comprised of 2D frequency fingerprints, which differ from the binary fingerprints since the frequency of specific atom pairs at different topological distances (from 1 to 10) is calculated.
The considered atom pairs are the combinations of C, N, O, S, P, F, Cl, Br, I, B, Si and X (where X represents any other atom, except hydrogen).
dragonX works only in background mode; moreover, it can support MDL, Sybyl, HyperChem, and SMILES input files. It has no restrictions on the molecule size (i.e. number of atoms) and the number of allowed molecules in a single batch. The only restriction is about the number of independent rings (cyclomatic number, nCIC) in a molecule that has been set to 19.
dragonX usage has been changed in order to be more compliant with Linux standards and the log has been improved.The main dragonX commands are the following:
Usage: dragonX [-h|-v|-c|-l [licensefile]|-d|-w|-s scriptfile]
-h, --help show usage message
-v, --version show version
-c, --citation show how to cite
-l, --license [licensefile] show license information (if licensefile is specified show license information of licensefile otherwise show license information on dragonX)
-d, --descriptors show the list of molecular descriptors
-w, --weights show the atom weights
-s, --script scriptfile run dragonX using scriptfile
In order to use dragonX with a script file using version 1.2 the following command line must be typed:
./dragonX script.txt [draghist.log]
where script.txt is a script file name and draghist.log is the file name for the log produced by dragonX; if the log file is omitted the logs are wrote to the standard output. Analogously, in order to run dragonX version 1.4 the command line should be:
./dragonX –s script.txt
The logs are redirected to the standard error; in order to capture them it is possible to use a command line like this:
./dragonX –s script.txt 2>draghist.log
in dragonX version 1.4 you can directly use the standard input to provide the molecules to dragonX; in order to do that the /fm line in the script file must be changed to
/fm <stdinput
dragonX ver. 1.4 can also write the results on the standard output; in order to do that the /fo line in the script file should be changed to
/fo >stdoutput
An example of the usage of dragonX with standard input and standard output options activated is:
cat moleculefile | dragonX script.txt >output.txt 2>draghist.log
Have a look to the script.txt file down here.
Another new feature introduced in dragonX version 1.4 is related to the script file; dragonX is able to interpret script files where not all the /d GetBnn lines are specified.
It means that the two following script files are interpreted in the same way:
DRAGON script Ver 2
/d GetB1 All /PCno
/d GetB2 All /PCno
/d GetB3 None /PCno
/d GetB4 None /PCno
/d GetB5 None /PCno
/d GetB6 None /PCno
/d GetB7 None /PCno
/d GetB9 None /PCno
/d GetB10 None /PCno
/d GetB11 None /PCno
/d GetB12 None /PCno
/d GetB13 All /PCno
/d GetB14 All /PCno
/d GetB15 All /PCno
/d GetB16 All /PCno
/d GetB17 All /PCno
/d GetB18 All /PCno
/d GetB19 All /PCno
/d GetB20 All /PCno
/d GetB21 None /PCno
/d GetB22 None /PCno
/fm <stdinput -f5 -i1 -Hy -3D
/fy None
/fo >stdoutput -f1 -k -m -999
DRAGON script Ver 2
/d GetB1 All /PCno
/d GetB2 All /PCno
/d GetB13 All /PCno
/d GetB14 All /PCno
/d GetB15 All /PCno
/d GetB16 All /PCno
/d GetB17 All /PCno
/d GetB18 All /PCno
/d GetB19 All /PCno
/d GetB20 All /PCno
/fm <stdinput -f5 -i1 -Hy -3D
/fy None
/fo >stdoutput -f1 -k -m -99