Kai Dührkop

SIRIUS 4.0 released

by

A new version of SIRIUS and CSI:FingerID is available for download. SIRIUS 4.0 supports structure elucidation for negative ion mode spectra and computes fragmentation trees up to 40 times faster. For positive ion mode spectra, CSI:FingerID shows an improved identification rate due to new training data and several methodical enhancements.
We also fixed a lot of bugs and think that CSI:FingerID is now working more stable.

You can download SIRIUS with CSI:FingerID here.

SIRIUS hotfix

by

We released a patch for SIRIUS+CSI:FingerID (now version 3.4.1) that fixes some critical bugs. We recommend everybody to download the newest version of SIRIUS.

SIRIUS+CSI:FingerID release

by

SIRIUS+CSI:FingerID leaves the alpha state. You can download it here. A few highlights of the new version:

  • Searching your tandem mass spectra in molecular databases using CSI:FingerID
  • Restrict the SIRIUS molecular formula identification to formulas appearing in molecular databases – or search through the whole space of possible formulas. It’s up to you.
  • Sort you identification results by confidence such that more reliable identifications (e.g. from high quality spectra or easy recognizable compounds) are separated from bogus identifications.
  • Predict structural features and substructures from tandem mass spectra and visualize them in your candidate structure list.

Don’t hesitate to give us your feedback, report bugs or suggesting new features. Just write to

SIRIUS+CSI:FingerID alpha release

by

The CASMI winner methods SIRIUS and CSI:FingerID are now available as a user interface. This allows users to batch process their spectra; which was not possible for the web application of CSI:FingerID.

You can try an early alpha version of SIRIUS+CSI:FingerID here.

Within the next three weeks we will release the final 3.1.4 version. Please report bugs to .

CSI:FingerID: Best automated method in CASMI contest

by

CSI:FingerID participated in this year’s CASMI (Critical Assessment of Small Molecule Identification) contest for automated methods identifying compounds solely by mass spectral data without additional meta data (category 2). IOKR, the new prediction method within CSI:FingerID, won this category. Standard CSI:FingerId prediction method ranked 2nd best. For positive ionization, CSI:FingerID identified more than twice as much compounds than any other method.

We are pleased that SIRIUS and CSI:FingerID were successfully used by many other contestants in category 1.

You can try CSI:FingerID on http://www.csi-fingerid.org/. A commandline version as well as a user interface which allows batch processing of MS/MS data will be released soon. We will also integrate the new IOKR prediction method into the CSI:FingerID web interface.

SIRIUS 3.0 release

by

We present SIRIUS 3.0, a java-based software for discovering a landscape of de-novo identification of metabolites using single and tandem mass spectrometry.
SIRIUS uses isotope pattern analysis for detecting the molecular formula and further analyses the fragmentation pattern of a compound using fragmentation trees.

The version 3.0 is a complete rewrite of our previous software. It uses faster algorithms for the fragmentation tree computation and a revised scoring based on Bayesian statistics.

A command line tool is available here. We will add a graphical user interface soon.

Lehrveranstaltungen von Ivo Große

by

Ivo Große von der Universität Halle wird uns am 9. und 16. Januar besuchen und zwei Vorlesungen zum Thema Motif Finding und EM-Algorithmus halten – klassische Themen der Bioinformatik, die in Jena leider nicht mehr gelehrt werden.

Die erste Vorlesung findet am Donnerstag dem 9. Januar von 13:00 bis 16:00 Uhr im Seminarraum 113, Carl-Zeiss-Str. 3 statt.

Die zweite Vorlesung findet am Donnerstag dem 16. Januar von 13:00 bis 16:00 im Hörsaal 146, UHG, Fürstengraben 1 statt.

Die beiden Veranstaltungen sind für alle Studierenden der Bioinformatik, insbesondere für alle Master-Studenten interessant. Alle Interessierten sind herzlich eingeladen.