Duration: 4-6 months
Location: Computational Systems Biology Team, LPHI, University of Montpellier
Many biological mechanisms rely on the synthesis, degradation, and activation of proteins. Among their numerous functions, proteins can act as signals, accelerating or reducing the production, degradation, or activation of other proteins,
allowing the studied system to adapt to external stimuli. These interactions form complex networks known as Protein Interaction Networks (PIN), the study of which is essential for understanding cell function in health and disease.
More formally, a PIN is a simple directed graph. However, the information it provides is static and insufficient to capture system dynamics. Consequently, Chemical Reaction Networks (CRNs) are often preferred. Unlike PINs,
CRNs are bipartite directed graphs that represent both proteins and the chemical reactions they participate in. A key advantage of CRNs is that they can be translated into systems of ordinary differential equations, allowing for the
modeling and simulation of a biological system's temporal evolution.
Since automating the transcription of a PIN into a CRN is not straightforward, the goal of this internship is to use a large collection of existing examples to train artificial intelligence algorithms for this task.
Representing PINs as networks naturally suggests using graph neural networks , a rapidly advancing architecture that captures the permutation invariance inherent to graphs. Since the goal is to produce a CRN,
which is also a network, the intern will study and apply available generative graph models (e.g., Variational Autoencoders, Attention Networks, Generative Adversarial Networks).
The solution developed during this internship will have applications in systems biology for medical research. As a proof of concept, we will apply the method to generate executable models of cellular signaling in inflammation.
This research is conducted in collaboration with King's College London, Cardiff University, the Center for Integrative Biology in Toulouse, and NCBS in Bangalore. During the internship, the student may benefit from a short stay in UK,
funded by the Sophie Germain program of the French Embassy in London.
This internship can be extended into a Ph.D. project.
Candidate profile: Theoretical and practical knowledge in machine learning and artificial intelligence is required, and an interest in biological applications would be appreciated.
Application: Send a CV + cover letter + transcripts + names of references to ovidiu.radulescu@umontpellier.fr
**********************************************************
*
* Contributions to be spread via DMANET are submitted to
*
* DMANET@zpr.uni-koeln.de
*
* Replies to a message carried on DMANET should NOT be
* addressed to DMANET but to the original sender. The
* original sender, however, is invited to prepare an
* update of the replies received and to communicate it
* via DMANET.
*
* DISCRETE MATHEMATICS AND ALGORITHMS NETWORK (DMANET)
* http://www.zaik.uni-koeln.de/AFS/publications/dmanet/
*
**********************************************************
Posts
