© Oré

ECML-PKDD 2022 workshop on machine learning for microbial genomics

We invite submissions to MLMG 2022, the ECML-PKDD 2022 workshop on machine learning for microbial genomics to be held virtually, on September 23rd.

The improvement and increasing accessibility of Next Generation Sequencing data calls for innovative data analysis strategies to extract meaningful and actionable information. Machine learning approaches are more and more popular in this context, offering alternative and complementary tools beyond bioinformatics to analyze genomic data. However, despite the recent advances, their application remains challenging and new approaches that are robust, scalable to large-scale high-dimensional data and interpretable are needed.

This workshop aims to bring together interdisciplinary researchers working at the interface of machine learning and computational biology for the study of microbial genomic data to discuss recent advances in this field. We will also have the privilege of hosting two distinguished speakers from this community to discuss and present their current research.

Typical topics of interest include:

- Improving microbial genome wide association studies (GWAS).
- Phenotype prediction from microbial genomes.
- Inferring population parameters from a set of microbial genomes.
- Visualization of microbial genomes in a way that highlights relevant
  elements with respect to a phenotype of interest.
- Study of the constitution of a microbial flora.

Key dates:

- Submission deadline: June 20th 2022 (instructions here)
- Decisions: July 13th 2022
- Online workshop: September 23rd

Meriem El Azami
Laurent Jacob
Flora Jay
Pierre Mahé


Lea Boulos joined the lab in February as a master 2 student. She works with Burak Yelmen, Guillaume Charpiat and I on generative models for large-scale genomic data - based on our previous work creating realistic genomes based on generative neural networks (GANs and RBMs). Lea's stipend is supported by ANR.

Lindsay Goulet also joined (in May) as a master 1 student working with Fanny Pouyet, Louis Ollivier and myself on demographic inference for yeast populations. Lindsay's stipend is upported by Fanny's grant.

DigiCosme Thematic School 2021 Graph as models in life sciences: Machine learning and integrative approaches - online - October 25th - 29th

We are organizing an online thematic school in bioinformatics and machine learning. You can register and spread the news.

Website: https://digicosme.cnrs.fr/ecole-thematique-2021-graph-as-models-in-life-sciences-machine-learning-and-integrative-approaches/

Registration form: https://framaforms.org/digicosme-thematic-school-2021-registration-1626095320
(free but mandatory, limited number of places for the tutorials)

 Flora Jay, CR CNRS, LISN
 Yann Ponty, DR CNRS, LIX
 Ariane Migault, Chargée de communication du Labex DigiCosme

 Burak Yelmen is joining the lab as a postdoctoral fellow to extend our previous work on creating realistic genomes based on generative neural networks (GANs and RBMs). Burak's salary is supported by ANR.

Pierre Jobic is doing a master internship in the lab and exploring deep architectures for population size inference. It will lead to a potentially substantial extension of our work with Théophile Sanchez, Jean Cury and Guillaume Charpiat :

T Sanchez, J Cury, G Charpiat, F Jay (2020). Deep learning for population size history inference: design, comparison and combination with approximate Bayesian computation. Molecular Ecology Ressources DOI:10.1111/1755-0998.13224 Link

Pierre is also helping  beta testing and improving our DNADNA software developed by the lab and in particular by Erik Madison Bray based on initial code by Théophile and Jean.
Mathieu Michel recently joined the lab for a master internship. He will help us benchmarking machine learning and deep learning methods for population genetics inference.