Join the team!

The Chair for Bioinformatics, Institute of Computer Science at the Friedrich-Schiller-University Jena offers the following position:

PostDoc/PhD students as a research associate (Wissenschaftliche Mitarbeiter)

The position is part of the project Harvester, funded by the Deutsche Forschungsgemeinschaft.

Topic of research: Metabolomics complements investigation of the genome, transcriptome, and proteome of an organism. Today, the vast majority of metabolites remain unknown. Mass spectrometry the two predominant experimental analysis techniques for detecting and identifying metabolites at small concentration. In metabolomics, natural products research and related areas, “identifying the unknowns” (small biomolecules for which no reference measurements are available) is of vital importance. Our method CSI:FingerID is currently the best-performing method for this task; our web services have processed half a billion queries for small molecule annotation. SIRIUS, CSI:FingerID and CANOPUS have been named “methods to watch” by Nature Methods twice; our latest tools CANOPUS and COSMIC both appeared in Nature Biotechnology.

Machine learning is an integral part of all of our recent methods. Unfortunately, training data for our methods are growing at a slow speed, and it will take a decade or two to double the currently available labeled training data. To this end, we want to resort to semi-supervised learning, as this will allow us to harvest the millions of spectra from unknown compounds that are publicly available. We have established the compute power to harvest spectra at a repository scale, and we have developed a Kubernetes-based framework for distributed computing. We have also established methods that allow us to “soft-label” previously unlabeled data. What we want to do in this method is bring all of that together, so that a) we can provide the largest public spectral library of biomolecules, and b) further improve our machine learning methods with this data.

See here for a beautiful introduction to this field of research, here for some background information on natural products research, and here and here for related research questions (all in German).

We offer: Challenging and interesting questions from different areas of computer science and machine learning, a network of cooperation partners in method development and application of our methods.

We are looking for: Applicants for this project must have or be about to obtain a Master’s degree or a qualification equivalent to the German Diploma in bioinformatics, chemoinformatics, computer science, or mathematics. Proven interest in machine learning is a must. Experience with Kubernetes and distributed computing is very helpful. Introductory course level knowledge in biology and molecular biology and good programming skills are required.

Salary is according to Entgeltgruppe 13 TV-L, either a full position (for PostDocs) or 75% position (for PhD students)

Handicapped applicants will be given preference in case of equal qualifications.

Please send your application with the usual documents (CV, list of publications, copies of certificates) via email and ASAP to:

Friedrich-Schiller-Universität Jena
Institut für Informatik
Lehrstuhl für Bioinformatik
Ernst-Abbe-Platz 2
07743 Jena, Germany