Mots-Clés

#deeplearning #bayesian #statistics #transposableelments #cancer

Description

Research Context & Impact

Transposable elements constitute approximately half of mammalian genomes and have emerged as central players in cancer biology, immune regulation, and therapeutic development. Their repetitive nature creates a fundamental computational barrier: sequenced reads cannot be uniquely mapped to specific loci, forcing researchers to discard 5–30% of sequencing data or rely on family-level averages that obscure critical locus-specific regulatory dynamics.

This project addresses these challenges through an integrated computational and biological framework that develops advanced multi-read allocation algorithms leveraging artificial intelligence to achieve locus-level resolution at TEs. The therapeutic relevance is direct: our work will enable rational design of TE-targeted epigenetic interventions, refine TE-based biomarkers for cancer diagnosis and prognosis, and inspire new therapeutic strategies exploiting viral mimicry for cancer immunotherapy. The team actively collaborates with the Sylvester Comprehensive Cancer Center experimental laboratories.

Required Qualifications

• PhD in Bioinformatics, Computational Biology, or Computer Science with biological applications. Candidates whose doctoral work focused on deep learning methods and who have a strong interest in Epigenetics will also be considered.
• At least one publication in computational genomics or machine learning methods
• Strong programming skills in Python and/or R
• Experience with deep learning frameworks (PyTorch or TensorFlow)
• Ability to work autonomously while maintaining regular communication

Preferred Qualifications

• Proven skills in Snakemake pipeline development with Conda environments and/or containerization
• Understanding of Expectation-Maximization algorithms or Bayesian statistical methods
• Familiarity with transposable element biology and/or repeat annotation pipelines
• Track record of software tool development and open-source contributions

Key Responsibilities

• Lead implementation of multi-read allocation algorithms and AI model development
• Conduct comprehensive benchmarking across diverse datasets, organisms, and genomic contexts
• Develop, document, and release production-quality software packages
• Prepare first-author manuscripts for high-impact journals
• Present research at major conferences
• Coordinate with experimental collaborators for biological validation