Richard Bonneau

MIMOSA: Guiding De Novo Antibody Design with Natural Interaction Fingerprints

MIMOSA improves de novo antibody design by constraining generative models to known binder interfaces, achieving validated binders on targets where unconstrained methods fail.

GLM-Prior: a genomic language model for transferable sequence-derived priors in gene regulatory network inference

GLM-Prior is a genomic language model that generates sequence-derived priors for gene regulatory network inference, showing strong performance in mammals but limited success in yeast.

DyAb: sequence-based antibody design and property prediction in a low-data regime

DyAb predicts antibody properties and designs new sequences with high binding affinity using a low-data approach.

Distinct Cellular Phenotypes of Language and Executive Decline in Amyotrophic Lateral Sclerosis

The paper maps cognitive impairments in ALS by integrating spatial and single-nucleus transcriptomics, revealing distinct cellular and molecular patterns in language and executive dysfunction.

Moving from Association to Causation: Instrumental factor models for causal inference in high-dimensional multi-omics data

This paper introduces FactorIV, a causal inference method for high-dimensional multi-omics data, demonstrating its ability to recover causal effects in biological systems.

CDR Conformation Aware Antibody Sequence Design with ConformAb

ConformAb designs antibody sequences that maintain CDR canonical conformations, improving binding affinity without target-specific data.

Tissue and cellular spatiotemporal dynamics in colon aging

This study creates a spatial and cellular atlas of the colon across age groups using spatial transcriptomics and single nucleus RNA-seq, identifying age-related molecular gradients and multicellular programs.

Property Enhancer - a data efficient multi-objective approach for functional antibody optimization

Property Enhancer optimizes antibody properties using a data-efficient framework, achieving significant affinity gains in both in silico and in vitro tests.

Dynamic Gene Regulatory Network Inference with Interpretable, Biophysically-Motivated Neural ODEs

This paper introduces in-CAHOOTTS, a neural ODE framework for inferring gene regulatory networks from dynamic single-cell data, showing it can predict gene expression and identify regulators in yeast responses to rapamycin and cell cycle progression.

PlasmoFP: leveraging deep learning to predict protein function of uncharacterized proteins across the malaria parasite genus

PlasmoFP uses deep learning to predict protein function in Plasmodium species, reducing unannotated proteins from 15-59% to 3-28% and increasing fully annotated proteins from 7-42% to 36-68%.

deepNGS Navigator: Exploring antibody NGS datasets using deep contrastive learning

deepNGS Navigator uses deep contrastive learning to visualize and analyze antibody sequence datasets, revealing clonal relationships and evolutionary patterns.

Comparative genomics of the sexually transmitted parasite Trichomonas vaginalis reveals relaxed and convergent evolution and genes involved in spillover from birds to humans

This study compares the genomics of seven Trichomonas species, revealing convergent genome size expansions in human-infecting lineages and genes involved in host-switching from birds to humans.

Deep Evolutionary Fitness Inference for Variant Nomination from Directed Evolution

This paper introduces EVFI and DeepEVFI, machine learning methods that infer variant fitness from directed evolution data, identifying tighter binders missed by conventional approaches.

Tokenized and Continuous Embedding Compressions of Protein Sequence and Structure

The paper compresses protein sequence and structure representations, showing that compressed embeddings retain structural accuracy and enable downstream applications.

Genome-wide expression gradient estimation based on local pseudotime in single cell RNA sequencing

The paper introduces pseudovelocity, a method to estimate gene-specific RNA velocity using kNN-G and diffusion-based pseudotime, improving gene-based velocity estimation for downstream analysis.

Lab-in-the-loop therapeutic antibody design with deep learning

This paper introduces a machine learning framework for therapeutic antibody design that integrates computational methods with experimental validation, resulting in significantly improved binding affinity for multiple targets.

Controllable All-Atom Protein Generation with Latent Diffusion

PLAID generates controllable all-atom protein structures from sequences using a diffusion model trained on sequence data, enabling functional control through Gene Ontology and organism keywords.

Meta-analysis of the human gut microbiome uncovers shared and distinct microbial signatures between diseases

This study performs a meta-analysis of gut microbiome data across 11 diseases, identifying shared and distinct microbial signatures using machine learning and differential abundance analysis.

Cramming Protein Language Model Training in 24 GPU Hours

This paper trains a protein language model in 24 GPU hours, achieving performance comparable to a much larger model trained for 15,000x more GPU hours.

OpenFold: Retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization

The paper trains a new version of AlphaFold2, OpenFold, and evaluates its generalization ability and learning mechanisms.

Structure primed embedding on the transcription factor manifold enables transparent model architectures for gene regulatory network and latent activity inference

This paper introduces a deep learning framework for gene regulatory network inference and latent transcription factor activity estimation, showing improved performance over existing methods.

Learning sequence, structure, and function representations of proteins with language models

The paper introduces a framework using protein language models to learn sequence-structure-function representations and develops a retrieval system for large-scale protein annotation and function prediction.

Simultaneous estimation of gene regulatory network structure and RNA kinetics from single cell gene expression

This paper develops a deep learning model to estimate gene regulatory networks and RNA kinetics from single-cell RNA-seq data, showing improved recovery of known regulatory relationships and predictive power.

Comprehensive function annotation of metagenomes and microbial genomes using a deep learning-based method

The paper develops a metagenome analysis workflow combining de novo genome reconstruction, taxonomic profiling, and deep learning-based functional annotation, showing improved coverage over existing methods.

Automated Protein Function Description for NovelClass Discovery

This paper introduces a neural machine translation model to generate natural language descriptions of protein functions, enabling the discovery of novel functions beyond existing label spaces.

EquiFold: Protein Structure Prediction with a Novel Coarse-Grained Structure Representation

EquiFold predicts all-atom protein structures using an SE(3)-equivariant coarse-grained geometric representation without sequence alignments or language model embeddings, matching AlphaFold accuracy orders of magnitude faster in preliminary tests.

A Variational Inference Approach to Single-Cell Gene Regulatory Network Inference using Probabilistic Matrix Factorization

This paper introduces PMF-GRN, a variational inference method for single-cell GRN inference, showing improved accuracy and uncertainty estimates compared to existing methods.

NetTIME: a multitask and base-pair resolution framework for improved transcription factor binding site prediction

NetTIME improves TF binding site prediction using multitask learning and base-pair resolution, outperforming existing methods in supervised and transfer learning settings.

TM-Vec: template modeling vectors for fast homology detection and alignment

TM-Vec predicts structural similarity from sequence pairs to detect remote homologs, outperforming sequence alignment methods and matching structure-based alignment performance.

Predicting recognition between T cell receptors and epitopes using contextualized motifs

TCRconv predicts T-cell receptor-epitope recognition using a deep learning model and provides insights into T-cell dynamics in COVID-19 patients.