Jorge FERNANDEZ-DE-COSSIO-DIAZ
CNRS Junior Professor
Institut de Physique Théorique
Contact: jorge[dot]fdcd[at]icloud[dot]com
We will be opening positions to work at the intersection of statistical physics, neural networks, and biological sequences. Contact me if interested.
I have internship positions available each year. See an example internship proposal at this link, as illustration of the topics I'm interested in.
I can also sponsor fellowship applications (such as MSCA, HFSP, EMBO, ...), contact if interested.
An 1 year postdoctoral position is available in my team to work on computational models of artificial evolution for aptamer design, expected to start in November 2025.
Aptamers are often obtained through rounds of in vitro competitive selection (SELEX), starting from artificial libraries of large numbers (~10^15) of distinct sequences. When the experiment succeeds, variants that perform the selected function well are enriched from one round to the next, and eventually the library collapses to a few selected functional aptamers. The evolution can be tracked in detail by deep sequencing of the successive rounds. The goal of the project is to exploit such data to develop generative models for aptamer design.
The candidate is expected to have a strong background in machine learning and statistical physics, with a real interest for computational biology.
This work will be carried out in close collaboration with the experimental team of Frédéric Ducongé, MIRCen, at the Institut de Biologie François Jacob of CEA.
To apply: https://academicjobsonline.org/ajo/jobs/30189. If you have any questions, send me an email.
We invite early-career researchers to apply for the STARφ programme of Université Paris-Saclay, funded by the European Union under the Marie Skłodowska-Curie COFUND action. My team at the IPhT at CEA-Saclay is among the participating laboratories.
I can support strong candidates applying for Marie Sklodowska Curie Postdoctoral Fellowships in my team. Get in touch if interested.
An 18 months postdoctoral position is available in Paris-Saclay, starting in March 2025, to work with Jorge FERNANDEZ-DE-COSSIO-DIAZ (IPhT, CEA-Saclay) and Martin WEIGT (Sorbonne University). The project aims at using methods based in statistical physics and machine learning / artificial intelligence to unveil signatures of protein functional specificity encoded in amino acid sequences. While many standard modeling approaches of protein sequences are based on multiple-sequence alignments (MSA), functional specificity is frequently encoded in hardly-alignable regions of variable length, and therefore discarded in standard sequence models. The goal of our project is to develop principled approaches able to handle variable sequence length and amino-acid insertions, and to relate them to protein function. We will, in particular, explore hierarchical models for sequence specific insertions, and protein language models.
We are looking for candidates with a strong quantitative background in machine learning, statistical physics or bioinformatics. We expect a strong motivation for interdisciplinary work; prior experience in biology is not required but considered a plus.
If you are interested, please contact jorge.fdcd@ipht.fr and martin.weigt@sorbonne-universite.fr by email.
This position is no longer available.