ISCB Affiliated Group Logo

Your role
You will develop computational methods and models expanding Kipoi, a collaborative initiative to define standards and to foster reuse of trained models in genomics [1]. Kipoi builds on a 3-way collaboration with international partners (Stegle lab, EMBL Heidelberg, and Kundaje lab, Stanford). The Kipoi model repository at is increasingly used and extended by the research community.

The position is funded via the recently awarded research network MechML. Research topics include: expressive mathematical representations of RNA or protein encoded regulatory sequences, notably using deep learning approaches (e.g. [2]); development of integrative models of individual steps of gene expression; development of methodologies for interpretability of deep learning models, and for their application to the prediction of causal effects of genetic variants in rare or common diseases (e.g. [3]). We expect applications on large-scale public data and on unpublished datasets from experimental collaborators in biology (e.g. [4]) or medicine (e.g. [5]).

You are
Applicants must either hold a PhD in bioinformatics, or hold a PhD in physics, statistics, or applied mathematics with practical experience with deep learning methods and application to real world high-dimensional data. (S)he must have a proven publication record, interest for translational research, and can work independently and creatively. (S)he should have excellent communications skills and be able to articulate clearly the scientific and technical needs, set clear goals and work within an interdisciplinary setting.

We are
The Gagneur lab is a lively, international, and interdisciplinary computational biology group with a research focus on the genetic basis of gene regulation and its implication in diseases. We are located in the informatics department of the Technical University of Munich, one of the top ranked European universities. Our lab has strong links to other local scientists and institutions in biology and medicine. Munich offers an outstanding, dynamic, interactive and internationally oriented research environment. Munich, the 2018 “most livable city in the world” according to the urban magazine Monocle, and the proximity of the Alps provide an excellent quality of life.

The position is funded for three years with a salary according to the TV-L (German academic salary scale).

Applications including a cover letter, CV, and references must be sent to Julien Gagneur (Diese E-Mail-Adresse ist vor Spambots geschützt! Zur Anzeige muss JavaScript eingeschaltet sein!, cc: Diese E-Mail-Adresse ist vor Spambots geschützt! Zur Anzeige muss JavaScript eingeschaltet sein!) until Nov 30th 2018 referring to “Postdoc-mechML” in the title.


1. Avsec et al., Kipoi: accelerating the community exchange and reuse of predictive models for genomics, bioRxiv, 2018
2. Avsec et al., Modeling positional effects of regulatory sequences with spline transformations increases prediction accuracy of deep neural networks, Bioinformatics, 2017
3. Cheng et al., Modular modeling improves the predictions of genetic variant effects on splicing, bioRxiv, 2018 – winner model of the CAGI 2018 splicing challenge
4. Schwalb et al., TT-seq maps the human transient transcriptome, Science, 2016
5. Kremer et al., Genetic diagnosis of Mendelian disorders via RNA sequencing, Nature communs, 2017