[Spectroscopy Ninja] Spectroscopy Ninja
© 2016: Dr. Friedrich Menges
[SpectraGryph] Spectragryph
- optical spectroscopy software -
Share this page on: share on Linkedin!
Contact & Connect: academia.edu Tumblr: Haunted by spectra Facebook: Spectroscopy Ninja Flickr: just some pics LinkedIn: my professional self! Twitter: Spectroscopy Ninja Xing: another professional resume ResearchGate: my academic self send an email my Google Plus youtube channel for Spectroscopy Ninja

About Testimonials Download Licenses Contact News
File Transform Analyze Fluorescence/ EEM Evaluate Automate Identify Acquire

Functions of the Identify ribbon
=> Click onto a button for descriptions

This ribbon is for working together with the database tab, which allows building up a reference database, assign meta data, add reference spectra from all file formats, define search algorithms, load sample spectra, execute searches and create reports from the search results.
See also the new collection of spectral databases in Spectragryph native file format, ready to download and for use with the new spectral database feature.

New identification view

Create a new Identification view as database tab for spectral database search. Each database tab holds one database and consists of four subdivisions (select as vertical tabs on the left side):
1) Database Structure: enter a database name and default file pathes, give a description, define descriptor categories and list descriptor values. "Method" is a mandatory descriptor, containing measurement methods as values.
2) Database Content: listing of reference entries, set reference ID, assign descriptor values, display reference spectra, search for reference entries
3) Search definition: create search definitions , restrict searches to parts of the database, combine and weight search definitions for combined result assessment
4) Search Execution & Results: load sample spectra, select search definitions, review search results, display sample spectra and best hits from database. Save and reload search results, export results to Excel, create PDF search reports sample-wise.

Database Structure: default descriptor "method" + one user-defined descriptor ("origin"). Name and description entered by user. Database Content: list of loaded reference spectra below, reference ID edited, three reference spectra selected and shown in the plot above. Search Definition: with two defined searches, differing in the used x axis range. Combine searches by using the same search name. Search Results: after loading a sample spectrum, selecting a search type and executing the database search, an ordered reference hit list is shown and can be displayed in plot above.
  <jump back to top>

Database creation & configuration

A Spectragryph database consists of three files:
*.sgdd: contains the database definition (first vertical tab)
*.sgdr: contains the database content, the reference entries (second vertical tab)
*.sgds: contains the search definitions (third vertical tab)
Additionally, *.sgsr: contains search results consisting of sample spectra and a hit list (fourth vertical tab)

For creating a new database, start in the first vertical tab (Database Structure). Enter a database name and a database description into the edit box on the right side. Optionally, a default filepath for the database files and for the reference spectra's original files can be defined. To better describe and categorize the database's reference entries, add your own descriptors and define a range of allowed descriptor values. These can be any kind of alphanumeric content. Additional rows can be created by pressing CTRL+ArrowDown within a cell from the lowest row. When finished with this tab, save the database with the [Save Database...] button. Saved databases are loaded with the [Load Database...] button.

Next, add reference spectra (all file formats, including OMNIC database files *.lbt/ *.lbd) with the [Import Reference Spectra ...] button. This happens in the second vertical tab.(Database Content). For every loaded reference spectrum, the "method" descriptor gets filled automatically, the "name" column is equivalent to the legend text (can't be edited), the unique db ID is assigned and, depending on the choosen options a reference ID is assigned. The reference ID can be edited and helps to assign and classify reference entries. There is a multitude of options for creating the reference ID from the legend text, filename or foldername. The same reference ID can be used for multiple reference entries. This allows to group several spectra of the same reference material, even from different measurement techniques.
If there are user-defined descriptor categories, they are listed in columns right to the "method" column. For each reference spectrum, a descriptor value can be choosen from the cell's dropdown list. When creating new descriptor categories of descriptor values after loading reference spectra, these are only available here after saving the database once more.
For showing reference spectra in the above plot window, select them in the "show" column. Clicking the checkbox in the "show" column's first row, results in showing/ hiding all reference entries (not wise if there are thousands).
The columns for reference ID and descriptor values each can be filtered for certain content, in order to show only a certain part of the reference entries. The search bar at the bottom allows to search for any text in the grid above.
Now, use the third vertical tab (Search Definition) to create search definitions. Start with the "compare the" column and select an algorithm from the dropdown list. When nothing special is desired, try first the "spectral similarity" algorithm. It works with all kind of spectra (UV-VIS absorbance and transmittance, NIR, FTIR, Raman, ...) and is based on calculating a full-spectrum Pearson correlation coefficent. Depending on the algorithm, some options for restricting the x axis range are available. Select one from the "x axis range" column and enter the numbers on the right into the appropriate cells. All algorithms can be applied to the original spectrum, as well as to a smoothed spectrum and first and second derivatives. While executing searches, the same transformation is temporarily applied to the reference spectra. By choosing a dedicated descriptor value within the "method" column and all user-defined descriptor columns to the right, the search range can be restricted to matching reference entries.
Search definitions sharing the same name ("search name" column) are executed together and give one common result and rating value. To visualize such search groups, click the check box in the first-row, leftmost cell (see left). For search groups, the result impact of eachs earch definition can be weighted by changing the value in the "result impact" column.
Search rows can be removed by simply pressing DEL key. Additional rows can be created by pressing CTRL+ArrowDown within a cell from the lowest row. The [Load Search Definitions] allows to load a saved search definition (file type *.sgds) and add the search defnition to the current list of search definitons.
  <jump back to top>

Database Search Execution

The real searching happens in the fourth vertical tab (Search Execution & Results). For any sample spectrum to search, load it first with the [Load Sample Spectra...] button. It is selected by default for searching (checkbox in "search" column) and the first search definition from the list of search definitions is selected in the "search name" column, also by default. Both can be changed as needed. The sample ID can be edited. As with reference spectra, a "sample" can also consist of several spectra of one or multiple types, to cover cases where several spectra describe one sample. As with the reference ID, the creation of the sample ID can be directed by changing the options in the options dialog.
Execute the search with the [Execute Database Search] button. Searches are comparing the sample spectrum against all reference spectra according to the selected search definition, restricted by descriptor selection and spectral method (only reference spectra with same method type are considered). As output, an ordered list of search hits are shown, with the hit rating value and the descriptor values of the found reference entries. Search results can be saved with the [Save Search Results...]. the number of displayed hits can be changed on the fly from the options dialog, as well as the number of saved hits. With the [Load Search Results...], previously saved search results can be loaded again into the same grid. .
The sample spectrum together with the best reference hits can be displayed in the plot above by clicking the checkbox in the "show" column. The search bar at the bottom allows to search for any text in the grid above.

<jump back to top>

Search Report generation

Here you can transfer the search results to Excel, either as one Excel file per sample (the upper left button) or else all results into a big single file (the lower button). The created Excel files will be instantly opened into Excel for review, if the option was set in the options dialog.
The [Create PDF Search Report(s)] allows to create a configurable and layouted PDF file that can serve as printable report. In the options dialog (see left), company name, address and logo can be entered. This will make up the header of the Search report. It further contains information about the used database (name & description), information about the searched sample (sample ID and according spectrum names), and finally the search results with best hit, hit table and a plot of sample spectrum with best hits. One search report is created per sample ID, and get's created for all present samples. Created reports get opened in your systems PDF reader, if the option was set in the options dialog.
In the options dialog, the report's page format can also be set. Additionally, All described database search options available in the options dialog can be permanently set ("Keep options permanently") with the general options feature from the "File" menu, so that they are already set after the next software start..

<jump back to top>
© 2001-2017: Friedrich Menges. Last Change: 16 May, 2017
Privacy statement: this website needs no cookies, it will neither track nor analyze you. No Google Analytics. No paid advertising. No selling of user data. No fuss. No shit. Only useful content for Spectroscopy Ninjas.