Bernhard Y. Renard

Decoding protein language models: insights from embedding space analysis

This paper introduces a framework for comparing embedding spaces in molecular biology, demonstrating its utility in pre-filtering machine learning tasks and identifying differential patterns across embeddings.

usiGrabber: Automating the curation of proteomics spectra data at scale, making large datasets ready for use in machine learning systems

usiGrabber automates proteomics data curation from repositories, enabling large-scale dataset creation for machine learning, achieving comparable performance to manual curation.

A High-Quality Acetylation Dataset Reveals Modest Data Requirements for Transfer Learning to Identify Little Studied Post-Translational Modifications

The paper presents a high-quality acetylation dataset and shows that transfer learning can achieve good performance with minimal data for identifying PTMs.

A High-Confidence Atlas of Protein Methylation Enables AI-Driven Detection of Methylated Peptides

The paper creates a high-confidence human methylation atlas and develops a deep learning model for methylated peptide detection.

ProteoDUDes: Taxonomic profiling for metaproteomics with false positive reduction

ProteoDUDes improves taxonomic profiling in metaproteomics by reducing false positives, achieving higher accuracy in identifying active organisms in complex samples.

Foundation Model Enables Interpretable Open and Error-Tolerant Searching for Mass Spectrometry-Based Proteomics

The paper introduces yHydra, a foundation model for mass spectrometry-based proteomics that enables open and error-tolerant peptide and protein identification with improved performance.

A framework for peptide identification on commercial nanopore sequencing platforms

This paper presents a framework for peptide identification using commercial nanopore sequencing platforms, achieving single-amino-acid resolution and reliable performance across sequencing runs.

Selecting Synthetic Data for Successful Simulation-Based Transfer Learning in Dynamical Biological Systems

The paper demonstrates how to select synthetic data for transfer learning in dynamical biological systems, showing performance improvements of up to 92% in mean absolute error.

Leveraging Uncertainty Estimates for Drug Response Prediction in Cancer Cell Lines

This paper benchmarks uncertainty-aware models for drug response prediction in cancer cell lines, finding ensemble-based methods better detect distribution shifts while distributional models reduce prediction errors in high-confidence subsets.

CombDNF: Disease-specific drug combination predictions with network-based features on clinically validated data

CombDNF predicts disease-specific drug combinations using network-based features and clinically validated data, outperforming existing methods by 84% in AUPR across four diseases.

End-to-end simulation of nanopore sequencing signals with feed-forward transformers

The paper introduces seq2squiggle, a transformer-based model for simulating nanopore sequencing signals, which outperforms existing tools on real data.

HiTaC: a hierarchical taxonomic classifier for fungal ITS sequences compatible with QIIME2

HiTaC is a hierarchical machine learning model for fungal ITS sequence classification that outperforms existing methods on noisy data.

PEPerMINT: Peptide Abundance Imputation in Mass Spectrometry-based Proteomics using Graph Neural Networks

PEPerMINT imputes missing peptide abundance values in proteomics using graph neural networks, outperforming existing methods across multiple datasets and missingness scenarios.

Taxor: Fast and space-efficient taxonomic classification of long reads with hierarchicalinterleaved XOR filters

Taxor classifies long reads in metagenomics with high precision and reduced memory use, outperforming existing tools.

Nanopore adaptive sampling effectively enriches bacterial plasmids

This study evaluates nanopore adaptive sampling for enriching low-abundant bacterial plasmids, showing improved assembly quality and reduced sequencing time.

Motif Interactions Affect Post-Hoc Interpretability of Genomic Convolutional Neural Networks

The paper examines how motif interactions affect post-hoc interpretability of genomic CNNs and shows that interpretability methods can miss motifs when interactions are present.

Predicting Decision-Making Time for Diagnosis over NGS Cycles: An Interpretable Machine Learning Approach

This paper predicts decision-making time for diagnosis during NGS cycles using an interpretable machine learning model, achieving over 98% accuracy and reducing turnaround time.

Evaluating Tuning Strategies for Sequence Generation with Protein Language Models

This paper evaluates prompt tuning as a cost-efficient alternative to fine-tuning protein language models for sequence generation, finding it more efficient in resource usage but less effective than fine-tuning in sequence quality.

Multi cell line analysis of lysosomal proteomes reveals unique features and novel lysosomal proteins

The paper analyzes lysosomal proteomes across six cell lines to identify cell-type specific differences and novel lysosomal proteins.

MpoxRadar: a worldwide Mpox genomic surveillance dashboard

MpoxRadar tracks Mpox virus mutations globally through an interactive dashboard, allowing users to analyze mutation profiles and filter data by various parameters.

Contamination detection and microbiome exploration with GRIMER

GRIMER detects contamination in microbiome studies by integrating multiple evidence sources and generating an interactive dashboard for data exploration.

Current state, existing challenges, and promising progress for de novo sequencing and assembly of monoclonal antibodies

This study evaluates de novo sequencing tools for monoclonal antibodies, finding that deep learning methods improve recall but challenges remain in heavy chain sequencing.

PathoLive - Real-time pathogen identification from metagenomic Illumina datasets

PathoLive detects pathogens in real-time from metagenomic data, improving speed and accuracy by prioritizing clinically relevant hits over contaminants and irrelevant sequences.

Detecting DNA of novel fungal pathogens using ResNets and a curated fungi-hosts data collection

The paper curates a fungal host range dataset and uses deep learning to detect novel fungal pathogens from DNA sequences.

DrDimont: Explainable drug response predictionfrom differential analysis of multi-omics networks

DrDimont predicts differential drug response by integrating multi-omics networks and provides explainable results, outperforming methods based on differential protein expression or PageRank in breast cancer data.

AHLF: ad hoc learning of peptide fragmentation from mass spectra enables an interpretable detection of phosphorylated and cross-linked peptides

AHLF is a deep learning model trained on mass spectra to detect phosphorylated and cross-linked peptides with improved performance and interpretability.

Critical Assessment of Metagenome Interpretation - the second round of challenges

The paper evaluates metagenomic software tools using realistic datasets, highlighting improvements in assembly and taxon profiling while identifying remaining challenges in virus/archaea detection and reproducibility.

ReadBouncer: Precise and Scalable Adaptive Sampling for Nanopore Sequencing

ReadBouncer improves nanopore adaptive sampling by combining fast base calling with read classification using Interleaved Bloom Filters, enhancing sensitivity and specificity for low-abundance sequence enrichment.

Deep learning-based real-time detection of novel pathogens during sequencing

This paper develops a deep learning method for real-time detection of novel pathogens during sequencing, showing improved sensitivity and accuracy compared to existing approaches.

MegaGO: a fast yet powerful approach to assess functional similarity across meta-omics data sets

MegaGO assesses functional similarity between meta-omics datasets using semantic GO term similarity, offering fast and scalable analysis.