LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transfer

Abstract Longitudinal multi-view omics data offer unique insights into the temporal dynamics of individual-level physiology, which provides opportunities to advance personalized healthcare. However, the common occurrence of incomplete views makes extrapolation tasks difficult, and there is a lack of...

Full description

Saved in:
Bibliographic Details
Main Authors: Siyu Han, Shixiang Yu, Mengya Shi, Makoto Harada, Jianhong Ge, Jiesheng Lin, Cornelia Prehn, Agnese Petrera, Ying Li, Flora Sam, Giuseppe Matullo, Jerzy Adamski, Karsten Suhre, Christian Gieger, Stefanie M. Hauck, Christian Herder, Michael Roden, Francesco Paolo Casale, Na Cai, Annette Peters, Rui Wang-Sattler
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-58314-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850153610094051328
author Siyu Han
Shixiang Yu
Mengya Shi
Makoto Harada
Jianhong Ge
Jiesheng Lin
Cornelia Prehn
Agnese Petrera
Ying Li
Flora Sam
Giuseppe Matullo
Jerzy Adamski
Karsten Suhre
Christian Gieger
Stefanie M. Hauck
Christian Herder
Michael Roden
Francesco Paolo Casale
Na Cai
Annette Peters
Rui Wang-Sattler
author_facet Siyu Han
Shixiang Yu
Mengya Shi
Makoto Harada
Jianhong Ge
Jiesheng Lin
Cornelia Prehn
Agnese Petrera
Ying Li
Flora Sam
Giuseppe Matullo
Jerzy Adamski
Karsten Suhre
Christian Gieger
Stefanie M. Hauck
Christian Herder
Michael Roden
Francesco Paolo Casale
Na Cai
Annette Peters
Rui Wang-Sattler
author_sort Siyu Han
collection DOAJ
description Abstract Longitudinal multi-view omics data offer unique insights into the temporal dynamics of individual-level physiology, which provides opportunities to advance personalized healthcare. However, the common occurrence of incomplete views makes extrapolation tasks difficult, and there is a lack of tailored methods for this critical issue. Here, we introduce LEOPARD, an innovative approach specifically designed to complete missing views in multi-timepoint omics data. By disentangling longitudinal omics data into content and temporal representations, LEOPARD transfers the temporal knowledge to the omics-specific content, thereby completing missing views. The effectiveness of LEOPARD is validated on four real-world omics datasets constructed with data from the MGH COVID study and the KORA cohort, spanning periods from 3 days to 14 years. Compared to conventional imputation methods, such as missForest, PMM, GLMM, and cGAN, LEOPARD yields the most robust results across the benchmark datasets. LEOPARD-imputed data also achieve the highest agreement with observed data in our analyses for age-associated metabolites detection, estimated glomerular filtration rate-associated proteins identification, and chronic kidney disease prediction. Our work takes the first step toward a generalized treatment of missing views in longitudinal omics data, enabling comprehensive exploration of temporal dynamics and providing valuable insights into personalized healthcare.
format Article
id doaj-art-e4e92326b3bf41d6a466f35ae89242a9
institution OA Journals
issn 2041-1723
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-e4e92326b3bf41d6a466f35ae89242a92025-08-20T02:25:40ZengNature PortfolioNature Communications2041-17232025-04-0116112010.1038/s41467-025-58314-3LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transferSiyu Han0Shixiang Yu1Mengya Shi2Makoto Harada3Jianhong Ge4Jiesheng Lin5Cornelia Prehn6Agnese Petrera7Ying Li8Flora Sam9Giuseppe Matullo10Jerzy Adamski11Karsten Suhre12Christian Gieger13Stefanie M. Hauck14Christian Herder15Michael Roden16Francesco Paolo Casale17Na Cai18Annette Peters19Rui Wang-Sattler20Institute of Translational Genomics, Helmholtz Zentrum München, German Research Center for Environmental HealthInstitute of Translational Genomics, Helmholtz Zentrum München, German Research Center for Environmental HealthInstitute of Translational Genomics, Helmholtz Zentrum München, German Research Center for Environmental HealthInstitute of Translational Genomics, Helmholtz Zentrum München, German Research Center for Environmental HealthInstitute of Translational Genomics, Helmholtz Zentrum München, German Research Center for Environmental HealthInstitute of Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental HealthMetabolomics and Proteomics Core, Helmholtz Zentrum München, German Research Center for Environmental HealthMetabolomics and Proteomics Core, Helmholtz Zentrum München, German Research Center for Environmental HealthCollege of Computer Science and Technology, Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education, Jilin UniversityEli Lilly and Company, Lilly Corporate CenterGenomics Variation, Population Medicine and Complex Diseases Unit, Turin UniversityInstitute of Experimental Genetics, Helmholtz Zentrum München, German Research Center for Environmental HealthBioinformatics Core, Weill Cornell Medicine-Qatar, Education CityInstitute of Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental HealthMetabolomics and Proteomics Core, Helmholtz Zentrum München, German Research Center for Environmental HealthInstitute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University DüsseldorfInstitute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University DüsseldorfInstitute of AI for Health, Helmholtz Zentrum München, German Research Center for Environmental HealthTUM School of Medicine and Health, Technical University of MunichGerman Center for Diabetes Research (DZD), Partner NeuherbergInstitute of Translational Genomics, Helmholtz Zentrum München, German Research Center for Environmental HealthAbstract Longitudinal multi-view omics data offer unique insights into the temporal dynamics of individual-level physiology, which provides opportunities to advance personalized healthcare. However, the common occurrence of incomplete views makes extrapolation tasks difficult, and there is a lack of tailored methods for this critical issue. Here, we introduce LEOPARD, an innovative approach specifically designed to complete missing views in multi-timepoint omics data. By disentangling longitudinal omics data into content and temporal representations, LEOPARD transfers the temporal knowledge to the omics-specific content, thereby completing missing views. The effectiveness of LEOPARD is validated on four real-world omics datasets constructed with data from the MGH COVID study and the KORA cohort, spanning periods from 3 days to 14 years. Compared to conventional imputation methods, such as missForest, PMM, GLMM, and cGAN, LEOPARD yields the most robust results across the benchmark datasets. LEOPARD-imputed data also achieve the highest agreement with observed data in our analyses for age-associated metabolites detection, estimated glomerular filtration rate-associated proteins identification, and chronic kidney disease prediction. Our work takes the first step toward a generalized treatment of missing views in longitudinal omics data, enabling comprehensive exploration of temporal dynamics and providing valuable insights into personalized healthcare.https://doi.org/10.1038/s41467-025-58314-3
spellingShingle Siyu Han
Shixiang Yu
Mengya Shi
Makoto Harada
Jianhong Ge
Jiesheng Lin
Cornelia Prehn
Agnese Petrera
Ying Li
Flora Sam
Giuseppe Matullo
Jerzy Adamski
Karsten Suhre
Christian Gieger
Stefanie M. Hauck
Christian Herder
Michael Roden
Francesco Paolo Casale
Na Cai
Annette Peters
Rui Wang-Sattler
LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transfer
Nature Communications
title LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transfer
title_full LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transfer
title_fullStr LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transfer
title_full_unstemmed LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transfer
title_short LEOPARD: missing view completion for multi-timepoint omics data via representation disentanglement and temporal knowledge transfer
title_sort leopard missing view completion for multi timepoint omics data via representation disentanglement and temporal knowledge transfer
url https://doi.org/10.1038/s41467-025-58314-3
work_keys_str_mv AT siyuhan leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT shixiangyu leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT mengyashi leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT makotoharada leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT jianhongge leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT jieshenglin leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT corneliaprehn leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT agnesepetrera leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT yingli leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT florasam leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT giuseppematullo leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT jerzyadamski leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT karstensuhre leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT christiangieger leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT stefaniemhauck leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT christianherder leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT michaelroden leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT francescopaolocasale leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT nacai leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT annettepeters leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer
AT ruiwangsattler leopardmissingviewcompletionformultitimepointomicsdataviarepresentationdisentanglementandtemporalknowledgetransfer