Create database spectra

From MicrobeMS Wiki
Jump to: navigation, search

Introduction

Database spectra are the preferred type of database entries in MicrobeMS. Such datebase spectra constitute the spectral references of a microbial library. Ideally database spectra are obtained from multiple measurements of the bacterial isolate under study in order to cover the spectral variability contained in the spectra of the biological and technical replicates.

The concept of database spectra involves automated generation of peak tables from spectral measurement series in which information about peak positions (m/z), normalized peak intensities and peak frequencies are obtained.

Mass peak tables from unknown microbial strains are then compared by pattern matching algorithms against a library of database spectra. The results of the pattern matching analyses can be provided as a matching rank rank list, either in a text, or a HTML format. These reports are easy to interprete and allow microbial identification down to the species level.

The database spectrum in the strict sense is represented by a simple average spectrum with a special type of peak table. Peak tables of database spectra contain the relevant information for the identification algorithms. Unlike the peak tables obtained from single experimental mass spectra the peak tables of database spectra contain also the peak frequency information.

Producing database spectra

 1. Load the mass spectral data files via the load spectra (Bruker data file format), 
    import spectra from mzXML data, or the load MS multifile options of the File pulldown menu.
Screenshot of the window create db spectra
2. To produce a database spectrum select the respective mass spectra in the listbox at the
   top left corner (the listbox is labeled by spectral tags). To select multiple spectra 
   hold the <shift> key while selecting.
3. Start the procedure by selecting create database spectra --> from selection from 
   the Analysis pull down menu. Database spectra can be also created from labeled spectra,
   i.e. from spectra with a class assignment. Labelling, or class assignment, can be carried out 
   by selecting the appropriate spectra and choosing class assignements --> class X from 
   the Edit pulldown menu.
4. As a result database spectra are created from peak lists produced from series of pre-processed 
   mass spectra. The parameters used for spectral pre-processing and peak detection are stored in 
   the MicrobeMS configuration file settings.opt.
   When the procedure is finished a window entitled view/edit metadata and pre-processing 
   parameters comes up which offers the possibilty to inspect and modify the metadata information.
   Furthermore, a new file id in the format dbspec-DD-MON-YEAR-HR-MI-SE.XXX will be created and
   assigned.
5. An option to create series of database spectra in an automated way is available from the 
   option create database spectra --> db spec batch mode (Analysis pulldown menu).
   A window entitled create db spectra offers the possibilty to indicate indices of the 
   mass spectra to be used. For example, to create a database spectrum from spectra 1-8 and to create a 
   second database spectrum from spectra 12-14 indicate 1 8 in the first line and 12 14 in 
   the second line. Press start when finished.

Related topics