Elise Jorge - Developing a Differential Analysis Method for Hi-C Data
Hi-C data provide insights into the three-dimensional organization of the genome by measuring interactions between genomic positions along the chromatin. Elise’s research focuses on a novel differential analysis method to identify genomic regions with significant structural differences between biological conditions. The method integrates interaction-wise differential analysis, two-dimensional agglomerative clustering with neighbor constraints, and post-hoc inference.
Sara Fallet - A Novel Gene Set Analysis Method for Single-Cell RNA-seq Data
Single-cell RNA-seq (scRNA-seq) enables molecular studies at single-cell resolution, especially valuable in immunology. Sara introduces a new gene set analysis method tailored for scRNA-seq, overcoming challenges of statistical power and heterogeneity. This method uses conditional distribution functions without relying on distributional assumptions and supports complex experimental designs to test associations of gene sets with multiple variables while adjusting for covariates.
Daniela Corbetta - Confidence-Aware Cell Type Annotation in Single-Cell RNA-seq
Recent advancements in single-cell RNA sequencing technologies have yielded diverse datasets and robust annotated references for cell annotation. Daniela’s research addresses the lack of uncertainty quantification in current cell annotation methods by proposing a prediction set-based approach. Using conformal risk control, her method integrates graph-structured constraints from cell ontology to return prediction sets that reflect confidence. The approach ensures accurate inclusion of true labels with a user-defined probability and demonstrates its effectiveness on real single-cell data.
Mathis Deronzier (INSA Toulouse, IMT)