Antoine Szatkownik

Interested in patterns
Machine Learning – Computational Biology

[Github]

[Google Scholar]

[Linkedin]

[Personal gallery]

October 30th 1997 antoine.szatkownik@gmail.com www.github.com/antoineszatkownik/

Experiences

2022
(6 months)
Berlin, GE

Robert Koch Institute (RKI)
Master internship joint with LCQB
Subject: Development of a computational method exploiting the interplay between gene duplication and alternative splicing to learn about the determinants of molecular recognition specificity
Supervisors: Hugues Richard and Elodie Laine

2021
(4 months)
Paris, FR

Laboratory of Computational and Quantitative Biology (LCQB)
Master internship
Title: Discovery of protein specificity signatures in evolutionary splicing graphs
Supervisors: Elodie Laine and Hugues Richard

Education

2022 – 2025
Gif-sur-Yvette, FR

Université Paris-Saclay
PhD in Computer Science, Laboratoire Interdisciplinaire des Sciences du Numérique (LISN)
Title: Latent generative modeling and synthetic data evaluation in population genomics
Supervisors: Flora Jay and Guillaume Charpiat
Co-supervisors: Burak Yelmen and Cyril Furtlehner

2020 – 2022
Paris, FR

Sorbonne Université
Master, Specialization in Bioinformatics and Modeling (BIM)

2018 – 2020
Orsay, FR

Université Paris-Sud
Bachelor, Fundamentals and Applied Mathematics (MFA)

2016 – 2018
Montréal, CN

Université de Montréal
Bachelor, Bidisciplinary in Mathematics and Physics

Teaching

2023 – 2024
Orsay, FR

MVA ENS + CentraleSupelec + DSBA ESSEC
Course: Deep Learning in Practice - M2 level
Lecturer: Guillaume Charpiat
Position: Teaching assistant (24h per academic year)

2022 – 2024
Orsay, FR

Université Paris-Saclay
Course: Introduction to machine learning - 3rnd year BCompSc level
Lecturer: François Landes
Position: Teaching assistant (24h per academic year)

Publications

[S.1] Szatkownik, A., Decelle, A., Seoane, B., Béreux, N., Charpiat, G., Yelmen, B., Jay, F., Furtlehner, C. (2025). PRIVET: PRIVacy metric based on Extreme value Theory. arXiv. 2510.24233


[P.2] Szatkownik, A., Planche, L., Demeulle, M., Chambe, T., Ávila-Arcos, M.C., Huerta-Sanchez, E., Furtlehner, C., Charpiat, G., Jay, F., Yelmen, B. (2024). Diffusion-based artificial genomes and their usefulness for local ancestry inference. bioRxiv. DOI: 10.1101/2024.10.28.620648


[P.1] Szatkownik, A., Furtlehner, C., Charpiat, G., Yelmen, B., Jay, F. (2024). Latent generative modeling of long genetic sequences with GANs. bioRxiv. DOI: 10.1101/2024.08.07.607012


[C.1] Szatkownik, A., Furtlehner, C., Charpiat, G., Yelmen, B., Jay, F. (2024). Towards creating longer genetic sequences with GANs: Generation in principal component space. Proceedings of the 18th Machine Learning in Computational Biology meeting, in Proceedings of Machine Learning Research, 240:110–122


[J.2] Yelmen, B., Decelle, A., Boulos, L.L., Szatkownik, A., Furtlehner, C., Charpiat, G., Jay, F. (2023). Deep convolutional and conditional neural networks for large-scale genomic data generation. PLOS Computational Biology, Vol. 19, Issue 10: e1011584. DOI: 10.1371/journal.pcbi.1011584


[J.1] Szatkownik, A., Javier Zea, D., Richard, H., Laine, E. (2023). Building alternative splicing and evolution-aware sequence-structure maps for protein repeats. Journal of Structural Biology, DOI: 10.1016/j.jsb.2023.107997


C=CONFERENCE, J=JOURNAL, S=IN SUBMISSION, P=PREPRINT

Research activities

    TALKS

  • LEGEND: Machine Learning for Evolutionary Genomics Data, Aussois, France, 2025.
  • Journée anniversaire des 10 ans de Sciences Animales Paris-Saclay, INRAE, Jouy-en-Josas, France, 2025.
  • Séminaire TAU, LISN, Gif-sur-Yvette, France, 2025.
  • Symposium on Synthetic clinical data, Queen Mary’s DERI, London, United Kingdom, 2025.
  • Séminaire MaIAGE, INRAE, Jouy-en-Josas, France, 2025.
  • Human Frontier Science Program meeting (HFSP), online, 2024.
  • LEGEND: Machine Learning for Evolutionary Genomics Data, Heraklion, Greece, 2024.
  • Machine Learning for Genomics (LEGO), Lille, France, 2023.
  • Séminaire BioInfo, LISN, Gif-sur-Yvette, France, 2023.
  • Journée des Doctorants, LISN, Gif-sur-Yvette, France, 2023.
  • yr2biomodel : second edition of the Young Researchers’ Meeting at Sorbonne Université, Paris, France, 2022.

    POSTERS

  • Northern Lights Deep Learning (NLDL), Tromsø, Norway, 2024.
  • Machine Learning in Computational Biology (MLCB), Seattle, United States, 2023.
  • Junior Conference on Data Science and Engineering (JDSE), Orsay, France, 2023.
  • Alphy & AIEM, Grenoble, France, 2023.
  • JOBIM, Rennes, France, 2022.

    REVIEWING

  • MLCB (also part of scientific Program Committee for 2025)

    SUPERVISION

  • Ilan Aliouchouche, Master student, Université Paris-Saclay
  • Titouan Chambe and Maïwen Demeulle, Master student, Université Paris-Saclay

    SEMINARS

  • Co-organizer of LISN BioInfo’s team seminar

Softwares

  • PRIVET: algorithm to detect instances of memorization and overfitting across diverse data modalities, including settings with very high dimensionality and limited sample sizes.
  • Latent generative models and evaluation metrics for artificial genomes (SNP sequences): PCA-WGAN, PCA-DDPM, k-SNPs motifs.
  • ASPRING: algorithm for evolutionary splicing graph objects to detect similar exonic regions that are modulated in number and composition by alternative splicing (Co-developed by Diego Javier Zea).

Skills

  • Programming: Python, R, Bash, Slurm, Git
  • Languages: French (Native), English (Fluent)
  • Interests: Sculpture (clay, modeling paste, steel, welding), Painting (acrylic), Drawing (charcoal), Collage, Computer graphics, Cinema, Writing (lino, fanzine, poetry)