Lucas Arnoldt
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
Lucas Arnoldt has a total of 133 co-authors. Here are the top co-authors by number of shared publications:
- Benjamin Wild
Digital Health, Berlin Institute of Health at Charité - Universitätsmedizin Berlin, Germany
3 shared publications - David Baker
Department of Biochemistry, University of Washington, Seattle, WA, USA
3 shared publications - Fabian Joachim Theis
Wellcome Sanger Institute, Wellcome Genome Campus, Cambridge, UK
3 shared publications - Julius Upmeier zu Belzen
Digital Health Center, Berlin Institute of Health (BIH) and Charité University Medicine, Berlin, Germany
3 shared publications - Roland Eils
Digital Health Center, Berlin Institute of Health (BIH) and Charité University Medicine, Berlin, Germany
3 shared publications - Samuel W. K. Tipps
Department of Biochemistry, University of Washington, Seattle, USA
3 shared publications
Latest Publications
- Open Access
Circulating immune architecture across age, exposures and disease in ethnically diverse UK cohorts.
Human immune composition reflects genetic background and cumulative life-course exposures, yet how these influences shape immune architecture across populations and disease remains poorly understood. Here we integrate single-cell transcriptomics, matched clinical haematology and electronic health records from 5,079 individuals of European, British-Pakistani and British-Bangladeshi ancestries to quantify 26 circulating immune cell types and 62 transcriptional states. We map immune compositional variation associated with age, sex, adiposity and smoking, revealing distinct patterns across adulthood and associations between modifiable exposures and advanced immune age. Across common diseases, immune compositional signatures converge on recurrent inflammatory myeloid, adaptive-memory and interferon programmes, whose expression varies with age and population context. Cardiovascular therapies are associated with selective shifts in disease-associated immune states. Immune composition further discriminates common cardiometabolic diseases and resolves patients into endotypes with distinct metabolic and clinical profiles. These findings establish circulating immune architecture as an integrative phenotype linking life-course exposures, population context and disease, providing a framework for precision population immunology.
Open AccessCirculating immune architecture across age, exposures and disease in ethnically diverse UK cohorts.
Human immune composition reflects genetic background and cumulative life-course exposures, yet how these influences shape immune architecture across populations and disease remains poorly understood. Here we integrate single-cell transcriptomics, matched clinical haematology and electronic health records from 5,079 individuals of European, British-Pakistani and British-Bangladeshi ancestries to quantify 26 circulating immune cell types and 62 transcriptional states. We map immune compositional variation associated with age, sex, adiposity and smoking, revealing distinct patterns across adulthood and associations between modifiable exposures and advanced immune age. Across common diseases, immune compositional signatures converge on recurrent inflammatory myeloid, adaptive-memory and interferon programmes, whose expression varies with age and population context. Cardiovascular therapies are associated with selective shifts in disease-associated immune states. Immune composition further discriminates common cardiometabolic diseases and resolves patients into endotypes with distinct metabolic and clinical profiles. These findings establish circulating immune architecture as an integrative phenotype linking life-course exposures, population context and disease, providing a framework for precision population immunology.
- Open Access
Multimodal population-scale cell atlases define archetypes and extremes of human blood
The cellular composition of blood is fundamental to human physiology and is routinely measured to assess health and disease. Circulating immune cells show extensive heterogeneity, but the organising principles and extremes of variation across individuals remain undefined. Here, using single-cell transcriptomics across two biobank cohorts encompassing ~6,000 individuals, we show that human blood occupies recurrent archetypal states defined by coordinated cell composition patterns. We delineate three major archetypes spanning naive lymphoid, myeloid and cytotoxic composition, and two minor archetypes characterised by type I interferon-driven inflammation and expanded premalignant CD5+ B cells. We develop cell2clone to infer lymphocyte clonal expansions, revealing how reactive clonal processes drive the cytotoxic archetype and premalignant clonal expansion establishes the CD5+ B cell archetype, creating outlier states that uncover context-specific effects of germline variants. Clonal haematopoiesis and mosaic chromosomal alterations influence blood states in an archetype-dependent manner. The archetype framework links blood states to common disease risk and clinical blood counts and, through a machine-learning model ArcheTypist, classifies infection and cancer immunotherapy outcomes. This population-scale framework has implications for diagnostics, early disease detection, immunotherapy stratification and mechanistic insights into disease susceptibility and therapy response, providing an organising model for identifying compositional subtypes across human tissues.
Open AccessMultimodal population-scale cell atlases define archetypes and extremes of human blood
The cellular composition of blood is fundamental to human physiology and is routinely measured to assess health and disease. Circulating immune cells show extensive heterogeneity, but the organising principles and extremes of variation across individuals remain undefined. Here, using single-cell transcriptomics across two biobank cohorts encompassing ~6,000 individuals, we show that human blood occupies recurrent archetypal states defined by coordinated cell composition patterns. We delineate three major archetypes spanning naive lymphoid, myeloid and cytotoxic composition, and two minor archetypes characterised by type I interferon-driven inflammation and expanded premalignant CD5+ B cells. We develop cell2clone to infer lymphocyte clonal expansions, revealing how reactive clonal processes drive the cytotoxic archetype and premalignant clonal expansion establishes the CD5+ B cell archetype, creating outlier states that uncover context-specific effects of germline variants. Clonal haematopoiesis and mosaic chromosomal alterations influence blood states in an archetype-dependent manner. The archetype framework links blood states to common disease risk and clinical blood counts and, through a machine-learning model ArcheTypist, classifies infection and cancer immunotherapy outcomes. This population-scale framework has implications for diagnostics, early disease detection, immunotherapy stratification and mechanistic insights into disease susceptibility and therapy response, providing an organising model for identifying compositional subtypes across human tissues.
- Open Access
An interpretable omnigenic neural network architecture for the human genome
Genetic prediction of complex phenotypes typically relies on additive linear models, which scale well but cannot capture non-additive effects or deeply integrate molecular and clinical data. Domain-specific neural networks have driven advances in images, text, and other modalities, but genome-scale neural networks remain challenging because genotypes are sparse and high-dimensional, effective sample sizes are limited, and generic architectures lack interpretability. Here, we introduce the omnigenic neural network, a biologically structured architecture inspired by the omnigenic model of complex traits. The model learns hierarchical representations of biological processes, accommodates multimodal inputs, supports transfer learning, and enables multitask prediction. Models trained in the UK Biobank and evaluated in the All of Us cohort for ischemic heart disease, type 2 diabetes, and schizophrenia outperformed published PGS Catalog and PRS-CSx scores. A multitask model trained across 36 cardiovascular endpoints further outperformed corresponding single-phenotype models and baselines. The architecture provides systems-level interpretability by quantifying the contributions of biological processes, which were consistent with established disease mechanisms. It also captures non-linear interactions between variants. Analysis of these interactions using Integrated Hessians revealed patterns concordant with previously reported epistatic associations. Together, these findings establish the omnigenic neural network as a flexible framework for interpretable, multimodal, and multitask genomic prediction.
Open AccessAn interpretable omnigenic neural network architecture for the human genome
Genetic prediction of complex phenotypes typically relies on additive linear models, which scale well but cannot capture non-additive effects or deeply integrate molecular and clinical data. Domain-specific neural networks have driven advances in images, text, and other modalities, but genome-scale neural networks remain challenging because genotypes are sparse and high-dimensional, effective sample sizes are limited, and generic architectures lack interpretability. Here, we introduce the omnigenic neural network, a biologically structured architecture inspired by the omnigenic model of complex traits. The model learns hierarchical representations of biological processes, accommodates multimodal inputs, supports transfer learning, and enables multitask prediction. Models trained in the UK Biobank and evaluated in the All of Us cohort for ischemic heart disease, type 2 diabetes, and schizophrenia outperformed published PGS Catalog and PRS-CSx scores. A multitask model trained across 36 cardiovascular endpoints further outperformed corresponding single-phenotype models and baselines. The architecture provides systems-level interpretability by quantifying the contributions of biological processes, which were consistent with established disease mechanisms. It also captures non-linear interactions between variants. Analysis of these interactions using Integrated Hessians revealed patterns concordant with previously reported epistatic associations. Together, these findings establish the omnigenic neural network as a flexible framework for interpretable, multimodal, and multitask genomic prediction.
- Open Access
- 19 March 2026
- Published by Elsevier in The Lancet Digital Health
- Vol. 8 (3) , 100962
A predictive atlas of disease onset from retinal fundus photographs: a modelling study using data from population-based cohorts
- 19 March 2026
- Published by Elsevier in The Lancet Digital Health
- Vol. 8 (3) , 100962
Open AccessA predictive atlas of disease onset from retinal fundus photographs: a modelling study using data from population-based cohorts
- Open Access
Biologically Guided Variational Inference for Interpretable Multimodal Single-Cell Integration and Mechanistic Discovery
Multi-omics technologies allow for a detailed characterization of cell types and states across multiple omics layers, helping to identify features that differentiate biological conditions, such as chemical or CRISPR-based perturbations. However, current tools employing variational inference on single-cell datasets, including methods for paired and mosaic integration, transfer learning, and modality imputation, typically act as black boxes. This lack of interpretability makes it challenging to evaluate whether biological variation is preserved, which can compromise downstream analyses. Here, we introduce NetworkVI, a sparse deep generative model designed for the integration and interpretation of multimodal single-cell data. NetworkVI utilizes biological prior knowledge as an inductive bias, specifically it relies on gene-gene interactions inferred from topologically associated domains and structured ontologies like the Gene Ontology to aggregate gene embeddings to cell embeddings, enhancing the interpretability at the gene and subcellular level. While achieving state-of-the-art data integration, modality imputation, and cell label transfer via query-to-reference mapping benchmarks across bimodal and trimodal datasets, NetworkVI additionally excels in providing biologically meaningful modality- and cell type-specific interpretations. NetworkVI aids researchers in identifying associations between genes and biological processes and uncovers immune evasion mechanisms in a Perturb CITE-seq dataset of melanoma cells. NetworkVI will support researchers in interpreting cellular disease mechanisms, guiding biomarker discovery, and ultimately aiding the development of targeted therapies in large-scale single-cell multimodal atlases. NetworkVI is available at http://github.com/LArnoldt/networkVI .
Open AccessBiologically Guided Variational Inference for Interpretable Multimodal Single-Cell Integration and Mechanistic Discovery
Multi-omics technologies allow for a detailed characterization of cell types and states across multiple omics layers, helping to identify features that differentiate biological conditions, such as chemical or CRISPR-based perturbations. However, current tools employing variational inference on single-cell datasets, including methods for paired and mosaic integration, transfer learning, and modality imputation, typically act as black boxes. This lack of interpretability makes it challenging to evaluate whether biological variation is preserved, which can compromise downstream analyses. Here, we introduce NetworkVI, a sparse deep generative model designed for the integration and interpretation of multimodal single-cell data. NetworkVI utilizes biological prior knowledge as an inductive bias, specifically it relies on gene-gene interactions inferred from topologically associated domains and structured ontologies like the Gene Ontology to aggregate gene embeddings to cell embeddings, enhancing the interpretability at the gene and subcellular level. While achieving state-of-the-art data integration, modality imputation, and cell label transfer via query-to-reference mapping benchmarks across bimodal and trimodal datasets, NetworkVI additionally excels in providing biologically meaningful modality- and cell type-specific interpretations. NetworkVI aids researchers in identifying associations between genes and biological processes and uncovers immune evasion mechanisms in a Perturb CITE-seq dataset of melanoma cells. NetworkVI will support researchers in interpreting cellular disease mechanisms, guiding biomarker discovery, and ultimately aiding the development of targeted therapies in large-scale single-cell multimodal atlases. NetworkVI is available at http://github.com/LArnoldt/networkVI .
- Open Access
- 7 October 2024
- Published by Springer Nature in Nature Biotechnology
- Vol. 43 (8) , 1384
Publisher Correction: Multistate and functional protein design using RoseTTAFold sequence space diffusion
- 7 October 2024
- Published by Springer Nature in Nature Biotechnology
- Vol. 43 (8) , 1384
Open AccessPublisher Correction: Multistate and functional protein design using RoseTTAFold sequence space diffusion
- Top Cited PapersOpen Access
- 25 September 2024
- Published by Springer Nature in Nature Biotechnology
- Vol. 43 (8) , 1288-1298
Multistate and functional protein design using RoseTTAFold sequence space diffusion
Protein denoising diffusion probabilistic models are used for the de novo generation of protein backbones but are limited in their ability to guide generation of proteins with sequence-specific attributes and functional properties. To overcome this limitation, we developed ProteinGenerator (PG), a sequence space diffusion model based on RoseTTAFold that simultaneously generates protein sequences and structures. Beginning from a noised sequence representation, PG generates sequence and structure pairs by iterative denoising, guided by desired sequence and structural protein attributes. We designed thermostable proteins with varying amino acid compositions and internal sequence repeats and cage bioactive peptides, such as melittin. By averaging sequence logits between diffusion trajectories with distinct structural constraints, we designed multistate parent–child protein triples in which the same sequence folds to different supersecondary structures when intact in the parent versus split into two child domains. PG design trajectories can be guided by experimental sequence–activity data, providing a general approach for integrated computational and experimental optimization of protein function.
- 25 September 2024
- Published by Springer Nature in Nature Biotechnology
- Vol. 43 (8) , 1288-1298
Top Cited PapersOpen AccessMultistate and functional protein design using RoseTTAFold sequence space diffusion
Protein denoising diffusion probabilistic models are used for the de novo generation of protein backbones but are limited in their ability to guide generation of proteins with sequence-specific attributes and functional properties. To overcome this limitation, we developed ProteinGenerator (PG), a sequence space diffusion model based on RoseTTAFold that simultaneously generates protein sequences and structures. Beginning from a noised sequence representation, PG generates sequence and structure pairs by iterative denoising, guided by desired sequence and structural protein attributes. We designed thermostable proteins with varying amino acid compositions and internal sequence repeats and cage bioactive peptides, such as melittin. By averaging sequence logits between diffusion trajectories with distinct structural constraints, we designed multistate parent–child protein triples in which the same sequence folds to different supersecondary structures when intact in the parent versus split into two child domains. PG design trajectories can be guided by experimental sequence–activity data, providing a general approach for integrated computational and experimental optimization of protein function.
- Open Access
Joint Generation of Protein Sequence and Structure with RoseTTAFold Sequence Space Diffusion
Protein denoising diffusion probabilistic models (DDPMs) show great promise in the de novo generation of protein backbones but are limited in their inability to guide generation of proteins with sequence specific attributes and functional properties. To overcome this limitation, we develop ProteinGenerator, a sequence space diffusion model based on RoseTTAfold that simultaneously generates protein sequences and structures. Beginning from random amino acid sequences, our model generates sequence and structure pairs by iterative denoising, guided by any desired sequence and structural protein attributes. To explore the versatility of this approach, we designed proteins enriched for specific amino acids, with internal sequence repeats, with masked bioactive peptides, with state dependent structures, and with key sequence features of specific protein families. ProteinGenerator readily generates sequence-structure pairs satisfying the input conditioning (sequence and/or structural) criteria, and experimental validation showed that the designs were monomeric by size exclusion chromatography (SEC), had the desired secondary structure content by circular dichroism (CD), and were thermostable up to 95°C. By enabling the simultaneous optimization of both sequence and structure, ProteinGenerator allows for the design of functional proteins with specific sequence and structural attributes, and paves the way for protein function optimization by active learning on sequence-activity datasets.
Open AccessJoint Generation of Protein Sequence and Structure with RoseTTAFold Sequence Space Diffusion
Protein denoising diffusion probabilistic models (DDPMs) show great promise in the de novo generation of protein backbones but are limited in their inability to guide generation of proteins with sequence specific attributes and functional properties. To overcome this limitation, we develop ProteinGenerator, a sequence space diffusion model based on RoseTTAfold that simultaneously generates protein sequences and structures. Beginning from random amino acid sequences, our model generates sequence and structure pairs by iterative denoising, guided by any desired sequence and structural protein attributes. To explore the versatility of this approach, we designed proteins enriched for specific amino acids, with internal sequence repeats, with masked bioactive peptides, with state dependent structures, and with key sequence features of specific protein families. ProteinGenerator readily generates sequence-structure pairs satisfying the input conditioning (sequence and/or structural) criteria, and experimental validation showed that the designs were monomeric by size exclusion chromatography (SEC), had the desired secondary structure content by circular dichroism (CD), and were thermostable up to 95°C. By enabling the simultaneous optimization of both sequence and structure, ProteinGenerator allows for the design of functional proteins with specific sequence and structural attributes, and paves the way for protein function optimization by active learning on sequence-activity datasets.