Paul Geeleher

Lineage-mixed neurodevelopmental programs in pediatric sarcomas nominate tumor-selective surface antigen pairs.

This paper develops an AI framework to interpret single-cell RNA-seq data from pediatric solid tumors, identifying tumor-specific surface antigen pairs for targeted therapies.

Tabular foundation model predicts alternative lengthening of telomeres (ALT) and identifies SMARCAL1 as a target in ALT-driven cancers

The paper develops ALTitude, a machine learning model to predict ALT status from whole genome sequencing data, identifies SMARCAL1 as a therapeutic target in ALT-driven cancers.

Bone morphogenetic protein (BMP) signaling determines neuroblastoma cell fate and sensitivity to retinoic acid.

The paper investigates RA's limited effect on primary neuroblastoma tumors but efficacy in clearing bone marrow cells, identifying BMP signaling as a key determinant of cell fate and RA sensitivity.

Preclinical Pediatric Molecular Analysis for Therapy Choice (MATCH)

This paper establishes a diverse collection of preclinical models for pediatric solid tumors and demonstrates their utility in precision medicine trials and drug testing.

Random-effects meta-analysis of effect sizes as a unified framework for gene set analysis

This paper introduces a unified framework for gene set analysis that improves performance and interpretability by comparing effect size distributions between gene sets and subgroups.

Cell type identification in spatial transcriptomics data can be improved by leveraging cell-type-informative paired tissue images using a Bayesian probabilistic model.

Discovering novel long non-coding RNA predictors of anticancer drug sensitivity beyond protein-coding genes

The paper identifies lncRNAs as significant predictors of anticancer drug sensitivity, showing they can enhance drug response predictions beyond protein-coding genes.

Cancer eQTLs can be determined from heterogeneous tumor gene expression data by modeling variation in tumor purity

The paper identifies cancer-specific eQTLs by modeling genotype-tumor purity interactions in heterogeneous tumor gene expression data, showing most risk variants affect tumor microenvironment cells.