Olivier Gevaert

Digital pathological classification of peripheral neuroblastic tumors across multi-continent cohorts

The paper develops NB-PathAssist, a computational framework for automated classification of peripheral neuroblastic tumors using whole-slide histology images, achieving high accuracy in prognostic grouping.

Integration of MRI radiomics and germline genetics to predict the IDH mutation status of gliomas

This study integrates MRI radiomics and germline genetics to predict IDH mutation status in gliomas, improving classification accuracy and prognostic risk assessment.

Revealing cancer driver genes through integrative transcriptomic and epigenomic analyses with Moonlight

The paper introduces Moonlight2, an updated framework that integrates transcriptomic and epigenomic data to identify cancer driver genes, finding 33, 190, and 263 epigenetically driven genes in three cancer subtypes.

Response to anti-angiogenic therapy is affected by AIMP protein family activity in glioblastoma and lower-grade gliomas

The paper investigates the role of AIMP proteins in glioma angiogenesis and their predictive value for anti-angiogenic therapy response using multi-omics and single-cell data.

JIND-Multi: Leveraging Multiple Labeled Datasets for Automated Annotation of Single-Cell RNA and ATAC Data

JIND-Multi improves cell annotation in single-cell RNA and ATAC data by transferring labels across multiple datasets, reducing unclassified cells and maintaining accuracy.

Deep learning uncovers histological patterns of YAP1/TEAD activity related to disease aggressiveness in cancer patients.

This study uses deep learning to predict YAP1/TEAD activity from histology slides, linking histological patterns to disease aggressiveness in cancer patients.

Towards Digital Quantification of Ploidy from Pan-Cancer Digital Pathology Slides using Deep Learning

PloiViT is a transformer-based model that quantifies tumor ploidy from pathology slides, outperforming traditional methods and achieving low prediction error across multiple cancer types.

regionalpcs: improved discovery of DNA methylation associations with complex traits

The paper introduces regionalpcs, a method for summarizing DNA methylation patterns, showing improved sensitivity in detecting associations with Alzheimer's disease traits.

GeNNius: An ultrafast drug-target interaction inference method based on graph neural networks

GeNNius uses a graph neural network to predict drug-target interactions more efficiently and accurately than existing methods across multiple datasets.

Digital profiling of cancer transcriptomes from histology images with grouped vision attention

This paper introduces SEQUOIA, a transformer model that predicts cancer transcriptomes from histology images, achieving high accuracy in predicting gene expression and recurrence risk in breast cancer.

Synthetic whole-slide image tile generation with gene expression profiles infused deep generative models

This paper introduces RNA-GAN, a deep generative model that creates synthetic whole-slide image tiles using gene expression profiles, outperforming traditional GANs and showing imputation capabilities.

EpiMix: an integrative tool for epigenomic subtyping using DNA methylation

EpiMix identifies differential DNA methylation in large cohorts with higher sensitivity than existing methods, revealing novel epigenetic mechanisms in childhood food allergy and lung cancer.

RNA-to-image multi-cancer synthesis using cascaded diffusion models

This paper introduces RNA-Cascaded-Diffusion-Model to generate synthetic whole-slide image tiles from RNA-Seq data, improving machine learning model performance in low-data settings.

ImaGene: A web-based software platform for tumor radiogenomic evaluation and reporting

ImaGene is a web-based platform for radiogenomic analysis that correlates tumor imaging and omics data, identifies gene-imaging feature associations, and generates diagnostic reports.

A meta-learning approach for genomic survival analysis

This paper proposes a meta-learning framework for genomic survival analysis, showing it outperforms transfer learning with fewer samples and identifies important genes and pathways in cancer.

Benchmark of lncRNA Quantification for RNA-Seq of Cancer Samples

This paper benchmarks lncRNA quantification methods on cancer RNA-Seq data, finding that pseudoalignment methods with full transcriptome annotation outperform alignment-based methods.

Proteogenomic view of cancer epigenetics: the impact of DNA methylation on the cancer proteome

The paper introduces ProteoMix, a method using proteomic data to identify epigenetic driver genes, finding that DNA methylation effects are often buffered at the protein level and enriched for cancer-related processes and markers.

Comparison of single gene and module-based methods for modeling gene regulatory networks

This study compares module-based and single gene methods for gene regulatory network inference, finding that a variational Bayes approach outperforms existing methods in both simulated and real data.

Deep Learning with Multimodal Representation for Pancancer Prognosis Prediction

The paper develops a deep learning model that predicts cancer patient survival across 20 cancer types using multimodal data, achieving a C-index of 0.784.

Dropout-Enabled Ensemble Learning for Multi-Scale Biomedical Data

The paper combines CNN models for histopathology and radiology images using a Dropout-Enabled meta-classifier to improve diagnostic accuracy in cancer.

Deep Recurrent Attention Models for Histopathological Image Analysis

The paper introduces a deep recurrent attention model for histopathological image analysis, showing comparable performance to CNNs while processing fewer patches.

A deep learning framework for imputing missing values in genomic data

This paper introduces a deep learning framework called DAPL for imputing missing values in genomic data, showing improved performance and reduced computational cost compared to traditional methods.

Locoregional Radiogenomic Models Capture Gene Expression Heterogeneity in Glioblastoma

This paper develops locoregional radiogenomic models to capture gene expression heterogeneity in glioblastoma using multi-omics data fusion and image analysis.

Module analysis captures pancancer (epi)genetically deregulated cancer driver genes for smoking and antiviral response

This paper introduces AMARETTO, a multi-omics integration algorithm that identifies pancancer driver genes related to smoking and antiviral responses by analyzing eleven cancer types.