Ioannis Tsamardinos

The paper introduces a conceptual framework and bioinformatics pipeline for multi-omics data integration, validated on TCGA datasets to enhance feature identification compared to single-omics analysis.

The paper constructs low-dimensional latent feature spaces of the human transcriptome using dimensionality reduction techniques and evaluates their ability to reconstruct data and improve phenotypic predictions.

The paper introduces JADBIO, an AutoML system for omics data that produces predictive models and biosignatures, showing it can overcome overfitting in high-dimensional, low-sample data.

The paper shows ERCC1-XPF interacts with TOP IIβ to repair transcription-induced DNA breaks, explaining clinical heterogeneity in related disorders.

The paper presents STATegra, a multi-omics dataset of mouse B-cell differentiation combining 10 omics technologies to study dynamic cellular processes.

This study compares three feature selection algorithms for gene expression data, finding that gOMP outperforms LASSO and matches FBED in performance while being more computationally efficient.

The paper applies causal discovery methods to mass cytometry data of human immune cells to predict causal relationships between signaling proteins and their responses to perturbations.