Edit Description

Editing the description

After starting a database query and selecting an entry in the result list the dataset is diplayed in the edit window. In the header area the dataset name, the default project language and the URI of the current dataset are show (see picture below). Below the header area the description name may be edited in a text box. Changes of the description data, e.g. the name, will be visible in the header after saving the dataset. To check dataset changes, click on the button to open the history. By clicking button you may open a simplified HTML form to view and edit the description data.

The main description data may be accessed in the first Mainattributes tab, which is described below. Click on one of the other tabs in the picture below to go directly to the appropriate manual chapter.

img/DD_Description_Main

 

Main attributes tab

The main attributes tab provides in the upper area a field to enter additional description details and a wording for natural language support. The optional filed AID may keep an alpha-numeric alternate ID of the dataset. If used, it should be unambigious within the project, but this is not checked by the program. Click on to get an overview of the used keys. 

 

Scopes

The next section Description scopes allows specification of description scopes. This sections consists of the scope table with scope type and value and a edit section to modify a seleted scope entry or to enter e new scope. The scope types Geographic ares, Reference, Specimen, Observation and Taxon name allow input of a free-form text or reference to the corresponding DiversityWorkbench module or a web service. The connection to to a module or web service is done using the button . After selecting the entry in the referenced database the text will be taken over and the backround colour of the scope value field changes to yellow. A further editing of the text will not be possible (see below).

 

Additionally the edit section now shows the buttons to view the database link and  to remove the database link. In the latter case editing of the scope value will be possible again.

If a scope entry has been selected in the scope table, it can be modified in the edit section. If you want to enter a new scope value instead, press the button at the end of the edit line. The scope entry in the scope table will be deselected and a selection box for the scope type appears in the edit section (see below).

You may insert mutiple scope values with links to a DiversityWorkbench module or a webservice. Therfore enter a new scope value with scope type Geographic ares, Reference, Specimen, Observation and Taxon name (or select one, which is not yet linked). Besides the scope type dropdown box the button is shown, which allows insert of multiple scopes (see image below).

You may insert mutiple scope values with links to a DiversityWorkbench module or a webservice. Therfore enter a new scope value with scope type Geographic ares, Reference, Specimen, Observation and Taxon name (or select one, which is not yet linked). Besides the scope type dropdown box the button is shown, which allows insert of multiple scopes (see image below).

In the remote query panel you find the section Unit list, where you may add the currently selected query result by clicking the button. When you select an entry in the Unit list, you may remove it by clicking or view the data in a separate window by clicking . When you have collected all query results click the OK button and the scopes are inserted for the description (see below).

 

Project scopes

Besides the already mentioned scope types that allow input of a free text or reference to a database entry, the scope types Sex, Stage, Part and Other scope provide pre-defined scope values. The values can be administrated as described in the Editing theProject chapter. For each project a different set of values can be selected as “recommended” values. If for the current description project scope values are defined/assigned, they are shown in the right part of the description scopes as Project scopes (see below).

Selecting or deleting of a scope entry can be easily done by selecting clickin on the check box before the scope value. You may select rsp. deselect all scope values of a certain scope type (Sex, State, Part or Other scope) by clicking the check box of the category. If for a scope value a detailled description text has been stored, you may view it as bubble help by moving the mous cursor over the values.

If for a description a scope value is selected, that is not assigned to its project, the value ist show with yellow background (see below). Such a situation can occur, if the scope value has been removed for the project or if a description has been moved to a different project. 

 

Project

The section Project allows the selection of the project the description is assigned to. This adjustment has influence of the available values of scope types Sex, Stage and Other scope, the available descriptors and recommended modifier/frequency values and statistical measures of the description. Therefore this unambiguous adjustment might be seen as the “terminology project” of the description item. The access rights for a certain user to the projects are controlled with the login settings (see chapter Loginadministration). When you click on a project name, you may view some additional project data (see image below middle). 

     

If access to a certain project is restricted to “read only”, it is displayed with grey colour and symbol in the Project section (see pictures above). A “read only” project cannot be selected for the actual description. If the selected description has already been assigned to a “read only” project by another user with appropriate rights, this is indicated at the right bottom corner of the application window (see above below right).

 

Additional projects

In section Additional projects you may make the description accessible for additional projects. To assign a description to a project click on the  button. To remove it from a project, select the project from the list and click on the  button. Additional projects without write access are shown in the upper part of the window, projects with full access in the lower (or only) part (see image below). If you do not have any access to a project, it is shown as red text. Be aware that by entering additional projects you may grant editing rights for single descriptions to users that have no write access to the current project!

     

Be aware that by entering additional projects you may grant editing rights for dedicated descriptions to users that have no write access to the current project! If you have write access to the terminologyproject of a description item and granted access to another project, where you have read-only access, you nevertheless may withdraw these rights by removing it from the Additional projects.

   

 

Continue with:

Jan 14, 2025

Subsections of Edit Description

Edit Description Continuous View

Editing the description - Continuous view tab

The continuous view tab provides an alternate view on the descriptive data. In the Summary data only the descriptors are included where data have been entered. At the right side there is a tool strip. The displayed buttons will change depending on the selected entry. The descriptor types categorical, quantitive, text and sequence are indicated by different colours and icons as shown in the example at the bottom. By pressing the button on the tool strip this example may be hidden. Descriptors that are marked as mandatory are displayed with light red text, descriptors that are inapplicable are greyed out. Additionally coloured background indicated possible problems, e.g. if an inapplicability rule has been ignored or a not allowed descriptor has been used. A tool tip text shows the error or warning reason when the mouse curser is moved over the element.

By pressing the button the tree view may be expanded to display the contained categorical states, quantitative measures or texts. By pressing the button the tree will be collapsed to descriptor level.

To insert a descriptor press , to delete a descriptor select the entry and press . If only a single value shall be removed, select the entry and use the button. All these functions can alternatively be accessed by the context menu by a right-click on the tree node.

If for a descriptor resources, e.g. pictures, are avaialble, this is indicated by the background color of the descriptor icon. Categorical states with resources are marked with icon (see first entries in picture above). The resources may be accessed by a right-click on the tree node and selecting context menu item View resources (see picture above).

 

Structured descriptor view

If for the project a structured descriptor tree is defined, i.e. a descriptor tree that contains at least one descriptor tree node, the Descriptor tree: drop-down list is shown below the descriptor tree (see image below left). If a descriptor tree is selected, the descriptor tree node hierarchy will be included in the output and the descriptors will be arranged accordingly. The names of the descriptor tree nodes and the descriptors will be reduced by parts contained in the supriour hierarchy nodes (see image below right).  

If for a descriptor tree node resources are available, this is indicated by the grey background color of the descriptor tree icon. The resources may be accessed by a right-click on the tree node and selecting context menu item View resources. Additional descriptor tree node information, e.g. an optional abbreviation or details text can be viewed as bubble help by placing the mouse cursor over the item.  

 

Editing summary data

A click on opens an edit window at the right part of the tab. This window provides the same functions as available in the descriptor view tab. The picture below shows the resource linked to a categorical state that can be opened by doule-clicking the state ID in the Categorical states section. 

 

Continue with:

Jan 14, 2025

Edit Description Descriptor View

Editing the description - Descriptor view tab

The left part of the descriptor view tab shows all descriptors available for the description’s project. The entries in column “!” indicate some additional information about the descriptor:

  • Descriptor is referenced in the summary data (●)
  • Descriptor is referenced in the status data (○)
  • Descriptor is mandatory, values must be specified (!)
  • Descriptor is inapplicable because of other specified descriptors/categorical states (x)
  • Descriptor is exclusive, only one categorical states shall be selected (e)

If resources are linked to a descriptor, the entry is marked by a square (□) behind its sequency number and a tool tip appears when the mouse cursor is moved over the “No.” column (see below). By double-clicking the sequence number a “Media view” window can be opened to view the resources.

 

If for the project descriptor trees are defined, the Descriptor tree: drop-down list and Restricted check box are shown below the descriptor list (see image below left). If a descriptor tree is selected, the names of the descriptor tree node hierarchy is included in the descriptor names (see image below right). This feature is useful, if several descriptors with ambigious names exist the can now be distinguished. E.g. two descriptors “colour”, one at descriptor tree node “Leaf”, the other at descriptor tree node “Blossom” will become “Leaf colour” and “Blossom colour”. If the descriptor name already includes the hierarchy, the duplicate parts will be filtered. In the mentioned example the descriptor names “Leaf colour” and “Blossom colour” will not be changed, if the descriptor tree is selected.

If additionally the Restricted option is set, only the descriptors included in the selected descriptor tree are displayed (see image below right, missing descriptor number 14). So you have the option to create several descriptor trees that include only special aspects of the description’s property (e.g. morphology, molecular biology and so on) to get a clearly arranged descriptor list. 

 

Editing summary data

The right part of the descriptor view tab shows at the bottom the Status data values of the selected descriptor. It is possible to select several values for each descriptor. In the “Notes” column additional information concerning the selected data status may be entered. By a double click on the “Notes” entry a separate edit window can be opened.

Depending on the descriptor type you can find at the top either Categorical states (see large picture above), Statistical measures, Descriptive text or Molecular sequence (see pictures below). In the “Notes” columns additional information concerning the specific categorical state rsp. statistical measure may be entered. By a double click on the “Notes” entry a separate edit window can be opened.

  

To enter a statistical measure value click in the “Value” field and enter the value, the check mark in field “x” will be set automatically. If recommended modifier values are defined (see “Editing the Descriptor” and “Editing the Project”), a value may be selected from the drop down list “Modifier”. If no modifier values are available, the appropriate table column may be hidden. In the “Notes” field an additional text may be added. By double-clicking the “Notes” field the Extended texteditor can be opened.

The descriptive text may be edited diretly in the window, which shows the pure text without any formatting. In the notes field an additional text may be entered. By double-clicking the text fields the Extended text editor can be opened. 

The molecular sequence window displays in the status line below the edit area the sequence type (“Nucleotide” or “Protein”), the symbol length (“Len:”) in case of proteine sequences, the gap symbol (“Gap:”) if defined, the long text of the actual element and the position (“Pos:”) (see image below). In case of three-letter proteine symbols the position is the number of the amino acid symbol, not the character position. By pressing the TAB key the cursor jumps to the next symbol. 

When characters are entered in the edit area that are recognized as valid symbols, they will automatically be formatted. One-letter symbols are converted to upper case characters, three-letter symbols are converted according the “Xxx” schema, e.g. “Ala” for “Alanine”. The sybols for the nucleotide bases "Adenine", "Cytosine", “Guanine”, "Thymine" and "Uracil" are dispayed with different colors. Ambiguity symbols and gap symbols are shown in grey color. If the ambiguity symbols have not been enabled in the descriptor data, they will be displayed with yellow background. Symbols that have not been recognized are displayed with red background. 

The molecular sequence window offers some additional functions available in the sequence menu. It may be opened by moving the mouse cursor over the control header (“Molecular sequence”, see window below left) or from the context menu (right-click on the sequence edit area, see windows below right). With menu item Import you may read the seqeunce data from a file and insert them to the description. With menu item Export you may export the actual sequence data to a file.  

   

By double-clicking the molecular sequence area or from sequence menu Open → Sequence a separate edit window may be opened (see image below). On the top of the window there is the drop-down menu Insert symbol: where a value may be selected from a list. After selecting the value press the [ENTER] key or click on to insert the selected symbol at the cursor position. 

To select a categorical state set a check mark in field “x” with a mouse click. If recommended modifier or frequency values are defined (see Editing the Descriptor and Editing the Project), values may be selected from the drop down lists “Modifier” rsp. “Frequency”. If no modifier or frequency values are available, the appropriate table column may be hidden. In the “Notes” field an additional text may be added. By double-clicking the “Notes” field the Extended text editor can be opened. 

If a “Modifier” or “Frequency” value is selected, the multiple specification of a dedicated categorical state is possible. A tool tip offers to add a row by doouble-click, if mouse cursor is moved over the “State” field (left picture below, picture in the midddle after double click). Multiple specifications of one categorical state are only allowed if they have been assigned different modifier and frequency values. Inputs offending these restrictions will be ignored an a cell error will indicate the problem (right picture below).

   

If resources are linked to a categorical state a tool tip appears when the mouse cursor is moved over the ID (left picture below). By double-clicking the ID a “Media view” window can be opened to view the resources (right picture below).

 

If more than one resources have been assigned to one entry, a slider will be visible at the bottom of the “Media view” window. To switch through the available resources the slider can be moved using the mouse or the right and left arrow keys (see below).

 

 

Continue with:

Jan 14, 2025

Edit Description HTML

Edit description in HTML form

With this form you can easily edit the description data in an HTML form. After starting a description query and selecting an entry choose Edit -> Edit as HTML form … from the menu or click on the button at the right side of the main Diversity Descriptions window. A window as shown below will be opened.

 

When you move the mouse cursor over embedded pictures, they are slightly magnified, by clicking on the resource, you may show it in full size (see below). Use button  Home to return to the HTML form. 

 

With drop down button Options you may insert some additional input fields in the form, e.g. item details or notes. With drop down button Scope you may insert input fields for the vasiour scope types and drop down button Color lets you select the background color. All that changes come into affect, when you reopen the form or redraw the HTML form using the Reload button.

After clicking the Reload button you will be asked if you want to discard all changes. By selecting No the HTML form data will be written into the local data store, otherwise all inputs will be reset. To exit the form and save all changes in the local data store, click the OK button. To ignore your changes, click the Cancel button and you will be asked if you want to discard all changes. The HTML form utility works completely on the local data store. I.e. if you edited the description data in the main form of Diversity Descriptions and start the HTLM form editor without saving the data, all changes are only present in the application dataset. To store the data in the database, click the save button of the main form!

 

 

May 3, 2024

Edit Description Molecular Sequences

Editing the description - Molecular sequence symbols

The editor for sequence data recognizes predefined symbols for nucleotide and protein sequences according the IUPAC definitions.

 

Nucleic acid symbols

Symbol Name
A Adenine
C Cytosine
G Guanine
T Thymine
U Uracile
W Weak (A or T) 
S Strong (G or C)
M aMino (A or C)
K Keto (G or T)
R puRine (G or A)
Y pYrimidine (C or T) 
B not A (B comes after A) 
D not C (D comes after C)
H not G (H comes after G)
V not T (V comes after T and U)
N No idea (not a gap)

The symbols with grey background are ambiguity symbols. The difference between “N” and a gap symbol (usually “-”, but any other symbol may be defined in the descriptor) is that a gap symbol represents an unspecified number of unknown symbols but “N” stands for exatly one nucleic acid. 

 

Amino acid symbols

Name 1-letter sybmol 3-letter sybmol
Alanine A Ala
Arginine R Arg
Asparagine N Asn
Aspartic acid S Asp
Cysteine C Cys
Glutamic acid E Glu
Glutamine Q Gln
Glycine G Gly
Histidine H His
Isoleucine I Ile
Leucine Leu
Lysine K Lys
Methionine M Met
Phenylalanine F Phe
Proline P Pro
Serine S Ser
Threonine T Thr
Tryptophan W Trp
Tyrosine Y Tyr
Valine V Val
Selenocysteine U Sec
Pyrrolysine O Pyl
Asparagine or aspartic acid B Asx
Glutamine or glutamic acid Z Glx
Leucine or Isoleucine J Xle
Unspecified or unknown amino acid X Xaa

The symbols with grey background are ambiguity symbols. The difference between “X” rsp. “Xaa” and a gap symbol (e.g. “—”, but any other symbol may be defined in the descriptor) is that a gap symbol represents an unspecified number of unknown symbols but “X” rsp. “Xaa” stand for exatly one amino acid. The amino acids “Selenocysteine” and Pyrrolysine" are non-standard amino acids that only occur in certain species. 

 

Continue with:

Jan 14, 2025

Edit Description Resources

The resouce links tab allows assigment and inspection of resource links for the description. In the lower Resources part of the tab there are two tables. Ar the left the “Resource” table specifies the single resources, at the right the “Resource variant” table contains the associated URLs to images, video or sound resources.

To enter a new resource select the empty line at the end of the “Resource” table and click on the “Resource name” field. After entering the “Resource name” (leave the cell by pressing the TAB key) automatically a new value for the sequence number (“No.”) is initialized. The sequence number determines the display order in tables and may be changed manually. You may order the resource entries by clicking on the column header or by using the arrow buttons in the left tool strip ( and for up and down, and to shift the selected entry to top or bottom). After ordering the resources click button to renumber the entries starting with “1” and make the changes effective. In field “Ranking” you may enter a numeric value in range 0 … 10 expressing the suitability of the resource. In field “Role” you may select one of the offered values.

If you want to delete a resource, you have three choices:

  1. Select the resource and press the button in the left tool strip.
  2. Select the resource and press the Delete key on the keyboard.
  3. Enter the sequence number column (“No.”) and remove the value. After leaving the cell the entry will be deleted.

In any case you will be asked if you want to delete the data row, if resource links are present. 

To view or edit the additional resource data and copyright information press on the button besides the resource table. In field License: and a link to the license text in the web may be entered, in field Rights: you may enter a copyright text. By pressing the button a browser window can be opened to navigate to the license page in the web. Furthermore you may enter a detailled text (field Details:), the resource language (field Language:) and set the Display embedded flag (see picture below). If any additional resource data have been entered, this will be indicated by a square (□) behind the resource name’s sequence number (see picture above).

 

To enter a resource link, first select the entry in the “Resource” table, then select the empty line at the end of the “Resource variant table”. By clicking on the “URL” field a web address may be entered. By double-clicking the “URL” field a browser window opens to navigate to the resource. For each resource several URLs to resource variants with different quality levels, e.g. different resolution, may be inserted. For each entry in the URL table a different value of “Variant” must be selected.

As an alternative to a resource URL a colour may be spcified in the format “color://#rrggbb” where “rrggbb” specifies a hexadecimal colour code. The colour can simply be seleced by clicking the button besides the resource variant table.

To remove a resource variant, select the entry and press the button in the right tool strip or the Delete key on the keyboard. Alternatively you may delete the “URL” entry in the resource variant table. 

The button besides the resource variant table allows you to view a resource URL in the system browser.

The upper part of the resource links tab consists of a preview window on the left side and the complete list of resource links. By selecting one of the links the preview is shown on the preview window. If you press button in the preview window, the resource is opened in a separate viewer window.

At the bottom of the resource links tab the media data of a selected URL are displayed. These data can be edited by double-clicking the “Type” entry in the resource variant table, by clicking at one of the media data value fields or the button. The edit dialog provides the possibility to access the URL to get available data. Additionally the values may be edited manually.   

 

Copy description resources to a descriptor

If you have write access to descriptors of the currently selected description’s project, you may copy a selected resource URL to a descriptor. After clicking the button a window will open to select one of the accessible descriptors (see image below). Remark: To see the copied resources in the Descriptorview or the Continuous view, you will have to reconnect to the database. 

 

Continue with::

Jan 14, 2025

Edit Description Samples

Editing the description - Sample data tab

The sample data tab allows assigment and inspection of sampling events and associated sampling units for the description. In the upper left Sampling events part of the tab there is a list of the sampling events stored for the description. By selecting a sampling event entry the Event data are shown in the upper right part of the tab. Here you may edit the event name, a detailled description, the date and time rsp. a date and time span and geographic data. To insert a new sampling event click on the button in the Sampling events section, to delete the selected sampling event and all stored sampling unit data press the button (see image above). 

The geographic area may be linked to an entry of a DiversityGazetteers database (see section Module related entry). If latitiude and longituede have not yet been entered, the coordinates of the selected DiversityGazetteers entry will be inserted. By pressing the button a window to select the coordinates by Google maps will be opened. Field Geo. datum allows entering short text concerning the geodetic datum. If coordinates are entered using Google maps a remark that WGS84 coordinates are used will be inserted.

 

Editing sampling unit data

For each sampling events one or several Sampling units may be entered. A sampling unit represents a tuple of values that belong together, e.g. because they represent a single specimen. For each descriptor, which is represented by the table column, a value may be entered in a sampling unit, which builds a single table line. The background colour of each table column indicates the type of sample data: categorical, quantitative, text or sequence . If the number of sampling units is displayed in field Units of the Sampling units section in the lower part of the tab. If sampling units are present, the number is displayed with yellow background (see image below).

To view the unit data, select a Descriptor tree and choose the descriptor sequence range (From descriptorto descriptor) to restrict the number of columns displayed in the unit table. Finally press button and the sampling untis will be displayed (see image below). If you prefer to display the present sampling units automatically whenever you select a sampling event, click on the number behind Units. The background changes from yellow to red to indicate this mode. When you select a cell and resources are available for the selected descriptor column, the button will be shown. You may click it to view the descriptor resources. 

To insert a new sampling unit press button . By clicking on the free space at the beginning of a table line you mark the whole sampling unit (see picture below). Now you have the additional options to delete the unit by pressing button or to copy the data to a new sampling unit by pressing the button. Keep in mind that all unit data will be copied, even if you do not see all descriptor columns due to the display restrictions.

 

When you select the Sampling unit ID cell you may enter a collection specimen in the Unit specimen control below the units table. By clicking the button (see section Module related entry) you can link it to an entry in a DiversityCollection database. If the sampling unit is linked to a database entry in DiversityCollection, the text of the Unit specimen: control is shown with light yellow background. You may view the linked data by clicking the button or double clicking the Sampling unit ID cell.

To modify categorical sampling data enter the table cell and press the button within the cell. A control will be opened where you can select the categorical states and enter notes or modifier values (if defined) for each single state (see image below).

 

For all other sampling data the value can be entered directly in the table cell. Notes for the selected table cell are entered in the Notes: text box directly below the units table. For categorical, text and molecular sequence sample data a separate edit window can be opened by double-clicking the table cell. For molecular sequence data the tool strip buttons (import) and  (export) allow import and export from rsp. to dedicated file formats as described in sections Import sequencedata and Export sequence data.

 

Continue with:

Jan 14, 2025

Export Sequence

Editing the description - Export sequence data

With this form you can export data from the database to a file. Choose Export from the sequence menu to open the window for the export. Format specifies the file format, currently FASTA and FASTQ are supported.

 

In the Header settings you may specify if a new header shall be generated or if the notes shall be searched for a FASTA rsp. FASTQ header. In case of FASTA export you may specify in the Comments settings section if FASTA comment lines from the notes (starting with “;”) shall be inserted after the header, the whole notes shall be inserted as FASTA comments or if the notes shall be ignored. If Insert description comments is checked, two comment linsed wit the description and descriptor title will be generated.

In the Header preview section you may check and edit the resulting header, in the Sequence preview section the converted sequence in one-letter notation is displayed.

In case of FASTQ export the Comments settings are irrelevant. If the notes include a QUALITY STRING sections, it is inserted in the FASTQ output (see image below). If no quality string can be found or the Find FASTQ quality from notes has been unchecked, the lowest quaality value “!” is inserted.

 

Click on the button to select the file where the data shall be exported. By default the data are appended at the end of the selected file, check the Overwrite existing file option to overwrite it. Finally click the Save file button to write the data.

 

Continue with: