Portal de Periódicos da CAPES

A reference map of potential determinants for the human serum metabolome

2020; Nature Portfolio; Volume: 588; Issue: 7836 Linguagem: Inglês

10.1038/s41586-020-2896-2

1476-4687

Noam Bar, Tal Korem, Omer Weissbrod, David Zeevi, Daphna Rothschild, Sigal Leviatan, Noa Kosower, Maya Lotan‐Pompan, Adina Weinberger, Caroline Le Roy, Cristina Menni, Alessia Visconti, Mario Falchi, Tim D. Spector, Henrik Vestergaard, Manimozhiyan Arumugam, Torben Hansen, Kristine H. Allin, Tue H. Hansen, Mun‐Gwan Hong, Jochen M. Schwenk, Ragna S. Häussler, Matilda Dale, Toni Giorgino, Marianne Rodriquez, Mandy H. Perry, Rachel Nice, Timothy J. McDonald, Andrew T. Hattersley, Angus G. Jones, Ulrike Graefe‐Mody, Patrick Baum, Rolf Grempler, Cecilia Engel Thomas, Federico De Masi, Caroline Brorsson, Gianluca Mazzoni, Rosa Lundbye Allesøe, Simon Rasmussen, Valborg Guðmundsdóttir, Agnes Martine Nielsen, Karina Banasik, Konstantinos D. Tsirigos, Birgitte Nilsson, Helle K. Pedersen, Søren Brunak, Tugce Karaderi, Agnete Troen Lundgaard, Joachim Johansen, Ramneek Gupta, Peter Wad Sackett, J. Tillner, Thorsten Lehr, Nina Scherer, Christiane Dings, Iryna Sihinevich, Heather Loftus, Louise Cabrelli, Donna McEvoy, Andrea Mari, Roberto Bizzotto, Andrea Tura, Leen M. ‘t Hart, Koen F. Dekkers, Nienke van Leeuwen, Roderick C. Slieker, Femke Rutters, Joline W. J. Beulens, Giel Nijpels, Anitra D.M. Koopman, Sabine van Oort, Lenka Groeneveld, Leif Groop, Petra J. M. Elders, Ana Viñuela, Anna Ramisch, Emmanouil Dermitzakis, Beate Ehrhardt, Christopher Jennison, Philippe Froguel, Mickaël Canouil, Amélie Boneford, Ian McVittie, Dianne Wake, Francesca Frau, Hans‐Henrik Stærfeldt, Kofi P. Adragni, Melissa K. Thomas, Han Wu, Imre Pavo, Birgit Steckel-Hamann, Henrik S. Thomsen, Giuseppe N. Giordano, Hugo Fitipaldi, Martin Ridderstråle, Azra Kurbasic, Naeimeh Atabaki Pasdar, Hugo Pomares‐Millan, Pascal M. Mutie, Robert W. Koivula, Nicky McRobert, Mark I. McCarthy, Agata Wesolowska‐Andersen, Anubha Mahajan, Moustafa Abdalla, Juan Fernandez, Reinhard W. Holl, Alison Heggie, Harshal Deshmukh, Anita M. Hennige, Susanna Bianzano, Barbara Thorand, Sapna Sharma, Harald Grallert, Jonathan Adam, Martina Troll, Andreas Fritsche, Anita Hill, Claire E. Thorne, Michelle Hudson, Teemu Kuulasmaa, Jagadish Vangipurapu, Markku Laakso, Henna Cederberg, Tarja Kokkola, Yunlong Jiao, Stephen Gough, Neil Robertson, Hélène Verkindt, Violeta Raverdi, Robert Caïazzo, François Pattou, Margaret H. White, Louise A. Donnelly, Andrew Brown, Nicholette D. Palmer, David Davtian, Adem Y. Dawed, Ian Forgie, Ewan R. Pearson, Hartmut Ruetten, Petra Musholt, Jimmy D. Bell, E. Louise Thomas, Brandon Whitcher, Mark Haid, Claudia Nicolay, Miranda Mourby, Jane Kaye, Nisha Shah, Harriet Teare, Gary Frost, Bernd Jablonka, Mathias Uhlén, Rebeca Eriksen, Josef Korbinian Vogt, Avirup Dutta, Anna Jönsson, Line Engelbrechtsen, Annemette Forman, Nadja B. Søndertoft, Nathalie de Préville, Tania Baltauss, Mark Walker, Johann Gassenhuber, Maria Klintenberg, Margit Bergstrom, Jorge Ferrer, Jerzy Adamski, Paul W. Franks, Oluf Pedersen, Eran Segal,

Bioinformatics and Genomic Networks

The serum metabolome contains a plethora of biomarkers and causative agents of various diseases, some of which are endogenously produced and some that have been taken up from the environment1. The origins of specific compounds are known, including metabolites that are highly heritable2,3, or those that are influenced by the gut microbiome4, by lifestyle choices such as smoking5, or by diet6. However, the key determinants of most metabolites are still poorly understood. Here we measured the levels of 1,251 metabolites in serum samples from a unique and deeply phenotyped healthy human cohort of 491 individuals. We applied machine-learning algorithms to predict metabolite levels in held-out individuals on the basis of host genetics, gut microbiome, clinical parameters, diet, lifestyle and anthropometric measurements, and obtained statistically significant predictions for more than 76% of the profiled metabolites. Diet and microbiome had the strongest predictive power, and each explained hundreds of metabolites—in some cases, explaining more than 50% of the observed variance. We further validated microbiome-related predictions by showing a high replication rate in two geographically independent cohorts7,8 that were not available to us when we trained the algorithms. We used feature attribution analysis9 to reveal specific dietary and bacterial interactions. We further demonstrate that some of these interactions might be causal, as some metabolites that we predicted to be positively associated with bread were found to increase after a randomized clinical trial of bread intervention. Overall, our results reveal potential determinants of more than 800 metabolites, paving the way towards a mechanistic understanding of alterations in metabolites under different conditions and to designing interventions for manipulating the levels of circulating metabolites. The levels of 1,251 metabolites are measured in 475 phenotyped individuals, and machine-learning algorithms reveal that diet and the microbiome are the determinants with the strongest predictive power for the levels of these metabolites.