Willem Waegeman

A systematic assessment of single-cell language model configurations

This study evaluates single-cell language model configurations by systematically testing pre-training strategies and identifies best practices for improving performance.

Reliable Molecular Retrieval from Mass Spectra using Conformal Prediction

This paper applies conformal prediction to molecular retrieval from mass spectra, producing reliable, spectrum-specific candidate sets with controlled reliability.

Evaluation of out-of-distribution detection methods for data shifts in single-cell transcriptomics

This study evaluates out-of-distribution detection methods for identifying novel cell types and data shifts in single-cell transcriptomics using synthetic and real datasets.

Harnessing machine learning to predict antibiotic susceptibility in Pseudomonas aeruginosa biofilms

The paper uses machine learning to predict tobramycin susceptibility in Pseudomonas aeruginosa biofilms using multiple analytical methods, achieving high accuracy with MALDI-TOF MS and Raman spectroscopy.

Pre-trained Maldi Transformers improve MALDI-TOF MS-based prediction

The paper introduces Maldi Transformer, a pre-trained model for MALDI-TOF MS data that improves prediction tasks through self-supervised learning.

An antimicrobial drug recommender system using MALDI-TOF MS and dual-branch neural networks

This paper develops a dual-branch neural network to predict antimicrobial resistance profiles from MALDI-TOF MS data, improving treatment efficiency and enabling fine-tuning across clinical labs.

Uncertainty-aware single-cell annotation with a hierarchical reject option

This paper evaluates rejection strategies in single-cell annotation, finding partial rejection in hierarchical classifiers improves reliability without losing label information.

A Comparison of Embedding Aggregation Strategies in Drug-Target Interaction Prediction

This study compares three embedding aggregation strategies for drug-target interaction prediction using benchmark datasets, showing when each strategy is most effective.

CpG Transformer for imputation of single-cell methylomes

The paper introduces CpG Transformer, a model for imputing single-cell methylomes, achieving state-of-the-art performance on scBS-seq and scRRBS-seq datasets.

Novel transformer networks for improved sequence labeling in genomics

The paper introduces transformer networks for sequence labeling in genomics, achieving state-of-the-art performance on tasks like transcription start site annotation in E. coli.

Explainable Transformer Models for Functional Genomics in Prokaryotes

This paper introduces transformer-based models for prokaryotic genome annotation and shows that attention heads specialize in identifying DNA binding sites.

DeepRibo: precise gene annotation of prokaryotes using deep learning and ribosome profiling data

DeepRibo uses deep learning and ribosome profiling data to accurately annotate gene expression in prokaryotes, outperforming traditional methods.

Algebraic Shortcuts for Leave-One-Out Cross-Validation in Supervised Network Inference

The paper introduces algebraic shortcuts for leave-one-out cross-validation to evaluate kernel-based network inference methods efficiently.