Florence Rufflé

The paper compares machine learning models to predict mutation status in AML patients using RNA-seq data, finding complex models achieve high accuracy for NPM1, IDH1/IDH2, and FLT3-ITD mutations.

The paper introduces a k-mer-based method to quantify RNA sequences from large RNA-seq datasets, demonstrating its accuracy for variant RNA detection in cancer cell lines.

The paper develops a k-mer based transcriptomics analysis to distinguish NPM1-mutated from non-mutated AML patients, achieving over 95% accuracy in three scenarios.

The paper introduces Kmerator Suite, a set of tools for extracting k-mer signatures and inferring metadata from RNA-seq datasets, showing high accuracy in gene expression quantification and metadata discovery.

The paper analyzes public RNAseq data to identify lncRNAs specific to mesenchymal stem cells and proposes them as potential cell markers.