Ali Oghabian, Birgitta W. van der Kolk, Pekka Marttinen, Armand Valsesia, Dominique Langin, W. H. Saris, Arne Astrup, Ellen E. Blaak, Kirsi H. Pietiläinen, University of Helsinki, Computer Science Professors, Nestlé Institute of Health Sciences, Toulouse University Hospitals, Maastricht University, Novo Nordisk Fonden, Department of Computer Science, Aalto-yliopisto, and Aalto University
Funding Information: The following grant information was disclosed by the authors: European Commission, the Food Quality and Safety Priority of the Sixth Framework Program: FP6-2005-513946. Finnish Diabetes Research Foundation. Academy of Finland: 335443, 314383, 272376 and 266286. Sigrid Jusélius Foundation. Academy of Finland Center of Excellence in Research on Mitochondria, Metabolism and Disease (FinMIT): 272376. Finnish Medical Foundation. Gyllenberg Foundation. Novo Nordisk Foundation: NNF20OC0060547, NNF17OC0027232 and NNF10OC1013354. Gyllenberg Foundation. Finnish Diabetes Research Foundation. Finnish Foundation for Cardiovascular Research. Funding Information: The Diogenes study was supported by the European Commission, the Food Quality and Safety Priority of the Sixth Framework Program (FP6-2005-513946). Birgitta W. van der Kolk was supported by the Finnish Diabetes Research Foundation. Kirsi H. Pietiläinen was funded by the Academy of Finland (grant numbers 335443, 314383, 272376 and 266286), Sigrid Jusélius Foundation, the Academy of Finland Center of Excellence in Research on Mitochondria, Metabolism and Disease (FinMIT; grant number 272376), the Finnish Medical Foundation, the Gyllenberg Foundation, the Novo Nordisk Foundation (grant numbers NNF20OC0060547, NNF17OC0027232 and NNF10OC1013354), Gyllenberg Foundation, the Finnish Diabetes Research Foundation, the Finnish Foundation for Cardiovascular Research, Government Research Funds, the University of Helsinki and Helsinki University Hospital. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Funding Information: At the time of the study, Armand Valsesia was a full-time employee at Nestlé Institute of Health Sciences SA. W. H. Saris discloses the receipt of research support from several food companies such as Nestlé, DSM, Unilever, Nutrition et Sante and Danone as well as pharmaceutical companies including GSK, Novartis and Novo Nordisk; he is also an unpaid scientific advisor for the International Life Science Institute (ILSI) Europe. Arne Astrup discloses grants and personal fees received from Gelesis, USA; personal fees from Acino, Switzerland; BioCare Copenhagen, DK; Dutch Beer Institute, NL; Groupe Éthique et Santé, France; IKEA Food Scientific Health Advisory Board, SE; McCain Foods Limited, USA; Navamedic, DK; Novo Nordisk, DK; Pfizer, USA; Saniona, DK; Weight Watchers, USA & Zaluvida, Switzerland; and grants from DC-Ingredients Denmark, which lie beyond the scope of the work reported here. Ellen E. Blaak received financial support from food industry, such as DSM, Danone, Friesland Campina, Avebe and Sensus, partly within the context of public–private consortia and has received funding from pharmaceutical companies including Novartis. She is involved in several task forces and expert groups related to ILSI Europe.g interests. Publisher Copyright: Copyright 2023 Oghabian et al. Background: Weight loss effectively reduces cardiometabolic health risks among people with overweight and obesity, but inter-individual variability in weight loss maintenance is large. Here we studied whether baseline gene expression in subcutaneous adipose tissue predicts diet-induced weight loss success. Methods: Within the 8-month multicenter dietary intervention study DiOGenes, we classified a low weight-losers (low-WL) group and a high-WL group based on median weight loss percentage (9.9%) from 281 individuals. Using RNA sequencing, we identified the significantly differentially expressed genes between high-WL and low-WL at baseline and their enriched pathways. We used this information together with support vector machines with linear kernel to build classifier models that predict the weight loss classes. Results: Prediction models based on a selection of genes that are associated with the discovered pathways ‘lipid metabolism’ (max AUC = 0.74, 95% CI [0.62–0.86]) and ‘response to virus’ (max AUC = 0.72,95% CI [0.61–0.83]) predicted the weight-loss classes high-WL/low-WL significantly better than models based on randomly selected genes (P < 0.01). The performance of the models based on ‘response to virus’ genes is highly dependent on those genes that are also associated with lipid metabolism. Incorporation of baseline clinical factors into these models did not noticeably enhance the model performance in most of the runs. This study demonstrates that baseline adipose tissue gene expression data, together with supervised machine learning, facilitates the characterization of the determinants of successful weight loss.