...defining the field

The following command line options are legal:

• almond [-o <filename.alm>]|[[-c <commandfile>][-q]]

Almond program can be started simply typing its name in any shell window with no argument.

• almond

This will start the GUI (Graphic User Interface) of Almond. This GUI contains menu panes, dialog windows, etc... which can be used to access all the Almond functionality.

Advanced Users can use the command line options to accelerate the opening of data files.

• almond -o <filename.alm>

From version 3.0, Almond offers the possibility to run the program without GUI, by providing a command file that describes the structures or fields to import and the computations to be done. The syntax to run Almond using a command file is as follows:

• almond -c <commandfile> -q

The optional -q argument produces Almond to run in quiet mode, without producing any output on the screen. Otherwise, the program dumps to the standard output the messages that are usually presented in the screen when the GUI is activated.

The syntax of the command file is described in the next section.

COMMAND FILE SYNTAX

Command files are plain ASCII text organized in lines. Lines starting with character # and empty lines are ignored. The order of the lines is irrelevant.

The command file must provide at least the name and type of the input structures or fields used for the computations and the name and type of the output written to disk. For all the rest of the information, Almond has default values defined, that can be changed with the appropiate commands.

Correct commands in Almond command files are:

Commands should appear as written above. Capitalization do matters. There should be just one space in between the command and the arguments and in between the different words composing the name of the command. The use of tabs is discouraged.

Defaults are:

import series mol2

keep KONT no

shape no

shape scale 6

shape field 0

shape cutoff 0.0

grid spacing 0.5

probes DRY O N1

scaling raw

filter nodes 100

filter weight 50

almd yes 1.0

macc2 window 0.8

macc2 size 0

almond output.alm

COMMAND FILE EXAMPLES

import series mol2

keep KONT yes

grid spacing 0.5

output GOLPE set00.dat

input list set00.lst

filter nodes 150

filter weight 35

macc2 window 0.8

almond set00.alm

This command will read the mol2 files, whose names are listed in file set00.lst and process all of them, running first GRID with probes DRY, O and N1 (the default) and a grid spacing of 0.5 Å and then processing the output in Almond. The kont files will be kept in disk. During the filtering, a maximum of 150 grid nodes will be extracted and a relative weight of 35 % will be given to the field value. The size of the auto-correlograms will be adjusted automatically (the default) and all the distance ranges will have a size of 0.8 grid unit. The resulting descriptors will be saved both in almond format as set00.alm and in GOLPE format in file set00.dat

import fields kont

probes OH2 DRY

input file file00.kont

output ASCII file00.txt

This command will read the kont file file00.kont. The fields described inside are assumed to be of type OH2 and DRY, as defined by command probe. The file will be processed using default filtering, MACC2 and scaling parameters. The resulting Almond variables will be written in ASCII format to file file00.txt. The resulting descriptors will be saved in almond format in file output.alm (the default).

import series sdf

sdflabel MOLNAME

probes OH DRY

input file cpd32.sdf

output ASCII cpd32.txt

This command will read the file in format SDFile cpd32.sdf. The structures described inside will be read one by one and processed by GRID, using probes OH and DRY, and the default 0.5 Å grid spacing. The file will be processed using default filtering, MACC2 and scaling parameters. The resulting Almond variables will be written in ASCII format to file cpd32.txt. The resulting descriptors will be saved in almond format in file output.alm (the default).

COMMAND DESCRIPTION

input file <filename>

input list <listname>

These two commands define the input. The input can be either a file containing the actual structure/field (input file) or a file containing a list of files names (input list). The argument should be the name of an existing file of the type specified.

output GOLPE <filename>

output ASCII <filename>

These two commands define the output. The Almond variables can be written in either GOLPE or ASCII format as defined by the name of the command. The argument should be a correct file name, according to UNIX syntax rules.

probes [OH | OH2 | C3 | DRY | O | O:: | OC2 | N: | N1 | N1: | N1+ | F | CL | BR | NA+ | K+ | CA+2 | FE+2]

When importing structures, this command defines the probes used to run GRID. When importing fields, this command describes the probes used by GRID when these fields were generated. In the first case, the order of the probe names doesn't matter, it will be automatically changed to be in agreement with order of the GUI probe chooser. In the second case, Almond assumes that the name, the order and the number of probes is correct and it will not check the correctness of the information provided. For example; if the number of fields included in the file are smaller than the number of probes, results can be undefined. A maximum number of three probes is allowed.

The arguments of the command are any number of probe names from the above list. The command should have at least one argument and a maximum of three. No other probe names should be used. The probes names should be written in capitals, separated by a single space.

The default is:

probes DRY O N1

shape [ yes | no ]

Activate the shape probe if the argument yes is selected. The shape probe will not be activated if four probes are already defined with the command probes.

The defaults is:

shape no

shape scale [INTEGER_VALUE]

shape field [OH | OH2 | C3 | O | O:: | OC2 | N: | N1 | N1: | N1+ | F | CL | BR | NA+ | K+ | CA+2 | FE+2]

shape cutoff [REAL_VALUE]

Shape parameters can be set explicitely. The scale parameter defines the radius of the sphere used for curvature calculation. The argument is a single integer number between 3 and 10 Å. The field parameter must be one of the probe defined with the command probes (DRY is not accepted) . The cutoff parameter is used to filter out the grid nodes of the shape field that are nearby the filtered nodes of the GRID fields. The argument is a real number which is the radius (in Å) of the sphere of exclusion, the sphere being centered on each filetered node of the GRID field. A cutoff of 0.0 Å means no exclusion .

The defaults is:

shape scale 6

shape field 0

shape cutoff 0.0

scaling [ raw | base | norma ]

When scaling is raw, no transformation is performed on the data. If option base is specified, all the variables that take zero value for one or more compounds are removed. A block-wise normalization between 0.5 and 2 is performed when option norma is selected. Only one scaling transformation can be used at the same time.

The default is:

scaling raw

filter nodes [INTEGER_VALUE]

filter weight [INTEGER_VALUE]

These two commands define the filtering parameters, option nodes is used to set the maximum number of filtered grid point extracted for each MIF, whereas option weight define the relative weight of the field. If the weight is set to 100, the selection will be based only upon the values of the field. On the other hand, if it is set to 0, the selection will be based only on the distances of the nodes.

The defaults are:

filter nodes 100

filter weight 50

almd [yes|no] [REAL_VALUE]

When almd is set to yes, the program will expect a real value between 0.0 and 1.0. If almd is set to no, no real value must be provided. If almd is set to yes, the MIF values will be tagged with the atom contributing most to every single field node. ALMOND will then use this information to prevent the GRIND to be computed using couples of nodes belowing to the same atom (intra-group couples). The real value is used to tune the electrostatic contribution to decide wether an atom is contributiong the most or not to a certain node.

The defaults are:

almd yes 1.0

macc2 window [REAL_VALUE]

macc2 size [INTEGER_VALUE]

These two commands define the auto-correlogram parameters. Option window stand for the width of the smooth window (in grid spacing unit), that is to say the size of the ranges that divide the distance space. Smooth window width may be set between 0.1 and 10. The size of the correlogram defines the number of variables included in each one of the correlogram blocks. When set to 0, the size is adjusted automatically so it can accommodate all the variables needed for the series, otherwise, the number of variables included is truncated or extended to match the size entered by the User. Valid values are positive integers between 0 and 100.

The defaults are:

macc2 window 0.8

macc2 size 0

import series [mmol2 | mol2 | kout | sdf]

import fields [kont | dat]

These commands inform Almond that the input file or list contains structures (import series) of type multi-mol2, mol2, kout or SDFile, or fields (import fields) of type kont or GOLPE. Almond will not check for the correctness of the information provided, and therefore, if the format of the input file does not correspond with the format entered, an error will be obtained.

Please notice that mol2 and kout file describe one single structure, while SDFiles and mmol2 can describe one or, more often, many structures. While it is possible to import lists with many mol2 or kout files, it is not possible to import lists of mmol2 or SDFiles.

The argument is one of the keywords provided, separated from the command name by a single space.

The default is:

import series mol2

grid spacing [REAL_VALUE]

This command defines the grid spacing to apply in the GRID calculations performed by Almond. It is not required when importing fields.

The argument is a single real number between 0.1 and 1.0 Å.

The default is:

grid spacing 0.5

keep KONT [yes|no]

This command determines if the the kont files obtained during GRID computations have to be removed from disk or not. It is not required when importing fields.

The default is:

keep KONT no

almond <filename>

This command defines the name given to the Almond .alm file generated during computations. This file is always written and can be opened after the computations were completed using the GUI, in order to review or process the results.

The argument can be any correct file name, according to UNIX syntax rules, but it is advisable to use the extension .alm.

The default is:

almond output.alm

sdflabel [TEXT_STRING]

When importing series from SDFiles, the name given to the objects is expected to be provided under a field labeled appropriately. The name of this label can be provided using this command, i.e. the command

sdflabel MOLNAME

will make Almond to look for the object name in a field called "<MOLNAME>" within the SDFile.

If no sdflabel command is present, the program will assume the first line of the SDFile, for each structure, to be the name of the object. If not found Almond will, in any case, provide a object name based in sequential numbers.

This command is ignored if the format of the input file is other than SDFile format.