Laetitia Meng-Papaxanthos
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Laetitia Meng-Papaxanthos has a total of 111 co-authors. Here are the top co-authors by number of shared publications:
- Jean-Philippe Vert
Owkin, Paris, France
5 shared publications - Ran Zhang
University of Washington
4 shared publications - William Stafford Noble
Department of Genome Sciences, University of Washington, Seattle, WA, USA
4 shared publications - Alex Bateman
European Molecular Biology Laboratory, European Bioinformatics Institute, (EMBL-EBI)
3 shared publications - Antonina Andreeva
EMBL-EBI
3 shared publications Felipe Llinares-López
Google DeepMind , 14 Rue de Londres, Paris 75009
3 shared publications
Latest Publications
- Open Access
AlphaGenome 2 predicts regulatory activity and variant effects across cell states
Most genetic variants associated with human traits and diseases lie in non-coding DNA, where they can alter gene regulation in specific cell types and states. Single-cell atlases chart these states across tissues and development, but rarely profile enough donors to associate variants with expression directly. Deep learning models could bridge this gap by predicting variant effects from DNA sequence, yet they poorly distinguish related cell states and require fine-tuning for new datasets. Here we present AlphaGenome 2 (AG2), which substantially advances cell-type-specific prediction of regulatory activity and variant effects for gene expression and chromatin accessibility. Trained on organism-wide human and mouse single-cell atlases and bulk functional genomics data, AG2 takes as input a 1-Mb DNA sequence and a condition specifying a target tissue, cell type, or cell state. Conditioning enables AG2 to impute unmeasured epigenomic tracks and predict multimodal variant effects across cell types. Inferring this condition directly from gene expression extends AG2 to unseen datasets and cell states without fine-tuning. AG2 outperforms fine-tuned AlphaGenome 1 and specialized models trained on target datasets, doubling recall of causal single-cell eQTLs at 90% precision. AG2 also achieves state-of-the-art enhancer–gene linking and can pinpoint causal variants at complex trait loci. Across hundreds of human cell types, we generated an atlas of enhancer–gene links and scores for 7 million common variants. As single-cell profiling expands across development, tissues, and disease cohorts, AG2 provides a blueprint for turning observational atlases into variant effect maps.
Open AccessAlphaGenome 2 predicts regulatory activity and variant effects across cell states
Most genetic variants associated with human traits and diseases lie in non-coding DNA, where they can alter gene regulation in specific cell types and states. Single-cell atlases chart these states across tissues and development, but rarely profile enough donors to associate variants with expression directly. Deep learning models could bridge this gap by predicting variant effects from DNA sequence, yet they poorly distinguish related cell states and require fine-tuning for new datasets. Here we present AlphaGenome 2 (AG2), which substantially advances cell-type-specific prediction of regulatory activity and variant effects for gene expression and chromatin accessibility. Trained on organism-wide human and mouse single-cell atlases and bulk functional genomics data, AG2 takes as input a 1-Mb DNA sequence and a condition specifying a target tissue, cell type, or cell state. Conditioning enables AG2 to impute unmeasured epigenomic tracks and predict multimodal variant effects across cell types. Inferring this condition directly from gene expression extends AG2 to unseen datasets and cell states without fine-tuning. AG2 outperforms fine-tuned AlphaGenome 1 and specialized models trained on target datasets, doubling recall of causal single-cell eQTLs at 90% precision. AG2 also achieves state-of-the-art enhancer–gene linking and can pinpoint causal variants at complex trait loci. Across hundreds of human cell types, we generated an atlas of enhancer–gene links and scores for 7 million common variants. As single-cell profiling expands across development, tissues, and disease cohorts, AG2 provides a blueprint for turning observational atlases into variant effect maps.
- Open Access
- 2 November 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
InterPro: the protein sequence classification resource in 2027
InterPro (https://www.ebi.ac.uk/interpro) is a freely accessible resource for the classification of protein sequences. It integrates predictive models, known as signatures, from multiple member databases to classify sequences into families and to predict the presence of domains and functionally important sites. The InterPro database provides annotations for hundreds of millions of sequences, ensuring extensive coverage of UniProtKB, the standard repository of protein sequences, and includes mapping to several other major resources, such as Gene Ontology, the Protein Data Bank in Europe (PDBe), and the AlphaFold Protein Structure Database. In this publication we report on the status of InterPro (version 110.0), detailing new developments in the database, its associated web interface, and software. Notable updates include InterPro-N, a deep-learning method that increases sequence coverage of UniProtKB to over 90%; the transition from InterProScan 5 to the completely re-engineered InterProScan 6; the integration of structure-based domain classifications and predicted structures from AlphaFold; a redesigned protein viewer offering a concise summary view; and a new Model Context Protocol (MCP) server that allows AI assistants to query InterPro programmatically. Over the past two years, 9484 new InterPro entries have been created, bringing the total to 54 922. InterPro data, software and tools are freely available.
- 2 November 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
Open AccessInterPro: the protein sequence classification resource in 2027
InterPro (https://www.ebi.ac.uk/interpro) is a freely accessible resource for the classification of protein sequences. It integrates predictive models, known as signatures, from multiple member databases to classify sequences into families and to predict the presence of domains and functionally important sites. The InterPro database provides annotations for hundreds of millions of sequences, ensuring extensive coverage of UniProtKB, the standard repository of protein sequences, and includes mapping to several other major resources, such as Gene Ontology, the Protein Data Bank in Europe (PDBe), and the AlphaFold Protein Structure Database. In this publication we report on the status of InterPro (version 110.0), detailing new developments in the database, its associated web interface, and software. Notable updates include InterPro-N, a deep-learning method that increases sequence coverage of UniProtKB to over 90%; the transition from InterProScan 5 to the completely re-engineered InterProScan 6; the integration of structure-based domain classifications and predicted structures from AlphaFold; a redesigned protein viewer offering a concise summary view; and a new Model Context Protocol (MCP) server that allows AI assistants to query InterPro programmatically. Over the past two years, 9484 new InterPro entries have been created, bringing the total to 54 922. InterPro data, software and tools are freely available.
- Open Access
- 16 October 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-14
Predicting sequence-specific amplification efficiency in multi-template PCR with deep learning
Multi-template polymerase chain reaction (PCR) is a critical technique enabling the parallel amplification of diverse DNA molecules, thereby facilitating applications in fields from quantitative molecular biology to DNA data storage. However, non-homogeneous amplification due to sequence-specific amplification efficiencies often results in skewed abundance data, compromising accuracy and sensitivity. In this study, we address amplification efficiency in complex amplicon libraries by employing one-dimensional convolutional neural networks (1D-CNNs) to predict sequence-specific amplification efficiencies, based on sequence information alone. Trained on reliably annotated datasets derived from synthetic DNA pools, these models achieve a high predictive performance (AUROC: 0.88, AUPRC: 0.44), thereby enabling the design of inherently homogeneous amplicon libraries. We further introduce CluMo, a deep learning interpretation framework that identifies specific motifs adjacent to adapter priming sites as closely associated with poor amplification. This insight leads to the elucidation of adapter-mediated self-priming as the major mechanism causing low amplification efficiency, challenging long-standing PCR design assumptions. By addressing the basis for non-homogeneous amplification in multi-template PCR, our deep-learning approach reduces the required sequencing depth to recover 99% of amplicon sequences fourfold, and opens new avenues to improve the efficiency of DNA amplification in fields such as genomics, diagnostics, and synthetic biology.
- 16 October 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-14
Open AccessPredicting sequence-specific amplification efficiency in multi-template PCR with deep learning
Multi-template polymerase chain reaction (PCR) is a critical technique enabling the parallel amplification of diverse DNA molecules, thereby facilitating applications in fields from quantitative molecular biology to DNA data storage. However, non-homogeneous amplification due to sequence-specific amplification efficiencies often results in skewed abundance data, compromising accuracy and sensitivity. In this study, we address amplification efficiency in complex amplicon libraries by employing one-dimensional convolutional neural networks (1D-CNNs) to predict sequence-specific amplification efficiencies, based on sequence information alone. Trained on reliably annotated datasets derived from synthetic DNA pools, these models achieve a high predictive performance (AUROC: 0.88, AUPRC: 0.44), thereby enabling the design of inherently homogeneous amplicon libraries. We further introduce CluMo, a deep learning interpretation framework that identifies specific motifs adjacent to adapter priming sites as closely associated with poor amplification. This insight leads to the elucidation of adapter-mediated self-priming as the major mechanism causing low amplification efficiency, challenging long-standing PCR design assumptions. By addressing the basis for non-homogeneous amplification in multi-template PCR, our deep-learning approach reduces the required sequencing depth to recover 99% of amplicon sequences fourfold, and opens new avenues to improve the efficiency of DNA amplification in fields such as genomics, diagnostics, and synthetic biology.
- Top Cited PapersOpen Access
- 20 November 2024
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 53 (D1) , D444-D456
InterPro: the protein sequence classification resource in 2025
InterPro (https://www.ebi.ac.uk/interpro) is a freely accessible resource for the classification of protein sequences into families. It integrates predictive models, known as signatures, from multiple member databases to classify sequences into families and predict the presence of domains and significant sites. The InterPro database provides annotations for over 200 million sequences, ensuring extensive coverage of UniProtKB, the standard repository of protein sequences, and includes mappings to several other major resources, such as Gene Ontology (GO), Protein Data Bank in Europe (PDBe) and the AlphaFold Protein Structure Database. In this publication, we report on the status of InterPro (version 101.0), detailing new developments in the database, associated web interface and software. Notable updates include the increased integration of structures predicted by AlphaFold and the enhanced description of protein families using artificial intelligence. Over the past two years, more than 5000 new InterPro entries have been created. The InterPro website now offers access to 85 000 protein families and domains from its member databases and serves as a long-term archive for retired databases. InterPro data, software and tools are freely available.
- 20 November 2024
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 53 (D1) , D444-D456
Top Cited PapersOpen AccessInterPro: the protein sequence classification resource in 2025
InterPro (https://www.ebi.ac.uk/interpro) is a freely accessible resource for the classification of protein sequences into families. It integrates predictive models, known as signatures, from multiple member databases to classify sequences into families and predict the presence of domains and significant sites. The InterPro database provides annotations for over 200 million sequences, ensuring extensive coverage of UniProtKB, the standard repository of protein sequences, and includes mappings to several other major resources, such as Gene Ontology (GO), Protein Data Bank in Europe (PDBe) and the AlphaFold Protein Structure Database. In this publication, we report on the status of InterPro (version 101.0), detailing new developments in the database, associated web interface and software. Notable updates include the increased integration of structures predicted by AlphaFold and the enhanced description of protein families using artificial intelligence. Over the past two years, more than 5000 new InterPro entries have been created. The InterPro website now offers access to 85 000 protein families and domains from its member databases and serves as a long-term archive for retired databases. InterPro data, software and tools are freely available.
- Top Cited PapersOpen Access
- 14 November 2024
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 53 (D1) , D523-D534
The Pfam protein families database: embracing AI/ML
The Pfam protein families database is a comprehensive collection of protein domains and families used for genome annotation and protein structure and function analysis (https://www.ebi.ac.uk/interpro/). This update describes major developments in Pfam since 2020, including decommissioning the Pfam website and integration with InterPro, harmonization with the ECOD structural classification, and expanded curation of metagenomic, microprotein and repeat-containing families. We highlight how AlphaFold structure predictions are being leveraged to refine domain boundaries and identify new domains. New families discovered through large-scale sequence similarity analysis of AlphaFold models are described. We also detail the development of Pfam-N, which uses deep learning to expand family coverage, achieving an 8.8% increase in UniProtKB coverage compared to standard Pfam. We discuss plans for more frequent Pfam releases integrated with InterPro and the potential for artificial intelligence to further assist curation. Despite recent advances, many protein families remain to be classified, and Pfam continues working toward comprehensive coverage of the protein universe.
- 14 November 2024
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 53 (D1) , D523-D534
Top Cited PapersOpen AccessThe Pfam protein families database: embracing AI/ML
The Pfam protein families database is a comprehensive collection of protein domains and families used for genome annotation and protein structure and function analysis (https://www.ebi.ac.uk/interpro/). This update describes major developments in Pfam since 2020, including decommissioning the Pfam website and integration with InterPro, harmonization with the ECOD structural classification, and expanded curation of metagenomic, microprotein and repeat-containing families. We highlight how AlphaFold structure predictions are being leveraged to refine domain boundaries and identify new domains. New families discovered through large-scale sequence similarity analysis of AlphaFold models are described. We also detail the development of Pfam-N, which uses deep learning to expand family coverage, achieving an 8.8% increase in UniProtKB coverage compared to standard Pfam. We discuss plans for more frequent Pfam releases integrated with InterPro and the potential for artificial intelligence to further assist curation. Despite recent advances, many protein families remain to be classified, and Pfam continues working toward comprehensive coverage of the protein universe.
- Open Access
- 8 July 2023
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 39 (7)
LSMMD-MA: scaling multimodal data integration for single-cell genomics data analysis
Motivation: Modality matching in single-cell omics data analysis—i.e. matching cells across datasets collected using different types of genomic assays—has become an important problem, because unifying perspectives across different technologies holds the promise of yielding biological and clinical discoveries. However, single-cell dataset sizes can now reach hundreds of thousands to millions of cells, which remain out of reach for most multimodal computational methods. Results: We propose LSMMD-MA, a large-scale Python implementation of the MMD-MA method for multimodal data integration. In LSMMD-MA, we reformulate the MMD-MA optimization problem using linear algebra and solve it with KeOps, a CUDA framework for symbolic matrix computation in Python. We show that LSMMD-MA scales to a million cells in each modality, two orders of magnitude greater than existing implementations. Availability and implementation: LSMMD-MA is freely available at https://github.com/google-research/large_scale_mmdma and archived at https://doi.org/10.5281/zenodo.8076311.
- 8 July 2023
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 39 (7)
Open AccessLSMMD-MA: scaling multimodal data integration for single-cell genomics data analysis
Motivation: Modality matching in single-cell omics data analysis—i.e. matching cells across datasets collected using different types of genomic assays—has become an important problem, because unifying perspectives across different technologies holds the promise of yielding biological and clinical discoveries. However, single-cell dataset sizes can now reach hundreds of thousands to millions of cells, which remain out of reach for most multimodal computational methods. Results: We propose LSMMD-MA, a large-scale Python implementation of the MMD-MA method for multimodal data integration. In LSMMD-MA, we reformulate the MMD-MA optimization problem using linear algebra and solve it with KeOps, a CUDA framework for symbolic matrix computation in Python. We show that LSMMD-MA scales to a million cells in each modality, two orders of magnitude greater than existing implementations. Availability and implementation: LSMMD-MA is freely available at https://github.com/google-research/large_scale_mmdma and archived at https://doi.org/10.5281/zenodo.8076311.
- Open Access
Mapping cells through time and space with moscot
Single-cell genomics technologies enable multimodal profiling of millions of cells across temporal and spatial dimensions. Experimental limitations prevent the measurement of all-encompassing cellular states in their native temporal dynamics or spatial tissue niche. Optimal transport theory has emerged as a powerful tool to overcome such constraints, enabling the recovery of the original cellular context. However, most algorithmic implementations currently available have not kept up the pace with increasing dataset complexity, so that current methods are unable to incorporate multimodal information or scale to single-cell atlases. Here, we introduce multi-omics single-cell optimal transport (moscot), a general and scalable framework for optimal transport applications in single-cell genomics, supporting multimodality across all applications. We demonstrate moscot’s ability to efficiently reconstruct developmental trajectories of 1.7 million cells of mouse embryos across 20 time points and identify driver genes for first heart field formation. The moscot formulation can be used to transport cells across spatial dimensions as well: To demonstrate this, we enrich spatial transcriptomics datasets by mapping multimodal information from single-cell profiles in a mouse liver sample, and align multiple coronal sections of the mouse brain. We then present moscot.spatiotemporal, a new approach that leverages gene expression across spatial and temporal dimensions to uncover the spatiotemporal dynamics of mouse embryogenesis. Finally, we disentangle lineage relationships in a novel murine, time-resolved pancreas development dataset using paired measurements of gene expression and chromatin accessibility, finding evidence for a shared ancestry between delta and epsilon cells. Moscot is available as an easy-to-use, open-source python package with extensive documentation at https://moscot-tools.org .
Open AccessMapping cells through time and space with moscot
Single-cell genomics technologies enable multimodal profiling of millions of cells across temporal and spatial dimensions. Experimental limitations prevent the measurement of all-encompassing cellular states in their native temporal dynamics or spatial tissue niche. Optimal transport theory has emerged as a powerful tool to overcome such constraints, enabling the recovery of the original cellular context. However, most algorithmic implementations currently available have not kept up the pace with increasing dataset complexity, so that current methods are unable to incorporate multimodal information or scale to single-cell atlases. Here, we introduce multi-omics single-cell optimal transport (moscot), a general and scalable framework for optimal transport applications in single-cell genomics, supporting multimodality across all applications. We demonstrate moscot’s ability to efficiently reconstruct developmental trajectories of 1.7 million cells of mouse embryos across 20 time points and identify driver genes for first heart field formation. The moscot formulation can be used to transport cells across spatial dimensions as well: To demonstrate this, we enrich spatial transcriptomics datasets by mapping multimodal information from single-cell profiles in a mouse liver sample, and align multiple coronal sections of the mouse brain. We then present moscot.spatiotemporal, a new approach that leverages gene expression across spatial and temporal dimensions to uncover the spatiotemporal dynamics of mouse embryogenesis. Finally, we disentangle lineage relationships in a novel murine, time-resolved pancreas development dataset using paired measurements of gene expression and chromatin accessibility, finding evidence for a shared ancestry between delta and epsilon cells. Moscot is available as an easy-to-use, open-source python package with extensive documentation at https://moscot-tools.org .
- Open Access
- 26 May 2022
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 38 (13) , 3454-3461
Conditional generative modeling forde novoprotein design with hierarchical functions
Protein design has become increasingly important for medical and biotechnological applications. Because of the complex mechanisms underlying protein formation, the creation of a novel protein requires tedious and time-consuming computational or experimental protocols. At the same time, machine learning has enabled the solving of complex problems by leveraging large amounts of available data, more recently with great improvements on the domain of generative modeling. Yet, generative models have mainly been applied to specific sub-problems of protein design. Here, we approach the problem of general-purpose protein design conditioned on functional labels of the hierarchical Gene Ontology. Since a canonical way to evaluate generative models in this domain is missing, we devise an evaluation scheme of several biologically and statistically inspired metrics. We then develop the conditional generative adversarial network ProteoGAN and show that it outperforms several classic and more recent deep-learning baselines for protein sequence generation. We further give insights into the model by analyzing hyperparameters and ablation baselines. Lastly, we hypothesize that a functionally conditional model could generate proteins with novel functions by combining labels and provide first steps into this direction of research. The code and data underlying this article are available on GitHub at https://github.com/timkucera/proteogan, and can be accessed with doi:10.5281/zenodo.6591379. Supplemental data are available at Bioinformatics online.
- 26 May 2022
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 38 (13) , 3454-3461
Open AccessConditional generative modeling forde novoprotein design with hierarchical functions
Protein design has become increasingly important for medical and biotechnological applications. Because of the complex mechanisms underlying protein formation, the creation of a novel protein requires tedious and time-consuming computational or experimental protocols. At the same time, machine learning has enabled the solving of complex problems by leveraging large amounts of available data, more recently with great improvements on the domain of generative modeling. Yet, generative models have mainly been applied to specific sub-problems of protein design. Here, we approach the problem of general-purpose protein design conditioned on functional labels of the hierarchical Gene Ontology. Since a canonical way to evaluate generative models in this domain is missing, we devise an evaluation scheme of several biologically and statistically inspired metrics. We then develop the conditional generative adversarial network ProteoGAN and show that it outperforms several classic and more recent deep-learning baselines for protein sequence generation. We further give insights into the model by analyzing hyperparameters and ablation baselines. Lastly, we hypothesize that a functionally conditional model could generate proteins with novel functions by combining labels and provide first steps into this direction of research. The code and data underlying this article are available on GitHub at https://github.com/timkucera/proteogan, and can be accessed with doi:10.5281/zenodo.6591379. Supplemental data are available at Bioinformatics online.
- 29 April 2022
- Published by Springer Nature
- p. 20-35
Semi-supervised Single-Cell Cross-modality Translation Using Polarbear
- 29 April 2022
- Published by Springer Nature
- p. 20-35
Semi-supervised Single-Cell Cross-modality Translation Using Polarbear
- Open Access
LSMMD-MA: Scaling multimodal data integration for single-cell genomics data analysis
Motivation Modality matching in single-cell omics data analysis—i.e., matching cells across data sets collected using different types of genomic assays—has become an important problem, because unifying perspectives across different technologies holds the promise of yielding biological and clinical discoveries. However, single-cell dataset sizes can now reach hundreds of thousands to millions of cells, which remains out of reach for most multi-modal computational methods. Results We propose LSMMD-MA, a large-scale Python implementation of the MMD-MA method for multimodal data integration. In LSMMD-MA we reformulate the MMD-MA optimization problem using linear algebra and solve it with KeOps, a CUDA framework for symbolic matrix computation in Python. We show that LSMMD-MA scales to a million cells in each modality, two orders of magnitude greater than existing implementations. Availability LSMMD-MA is freely available at https://github.com/google-research/large_scale_mmdma Contact lpapaxanthos{at}google.com
Open AccessLSMMD-MA: Scaling multimodal data integration for single-cell genomics data analysis
Motivation Modality matching in single-cell omics data analysis—i.e., matching cells across data sets collected using different types of genomic assays—has become an important problem, because unifying perspectives across different technologies holds the promise of yielding biological and clinical discoveries. However, single-cell dataset sizes can now reach hundreds of thousands to millions of cells, which remains out of reach for most multi-modal computational methods. Results We propose LSMMD-MA, a large-scale Python implementation of the MMD-MA method for multimodal data integration. In LSMMD-MA we reformulate the MMD-MA optimization problem using linear algebra and solve it with KeOps, a CUDA framework for symbolic matrix computation in Python. We show that LSMMD-MA scales to a million cells in each modality, two orders of magnitude greater than existing implementations. Availability LSMMD-MA is freely available at https://github.com/google-research/large_scale_mmdma Contact lpapaxanthos{at}google.com