Dataset Information

An integrative machine learning approach to discovering multi-level molecular mechanisms of obesity using data from monozygotic twin pairs.

ABSTRACT: We combined clinical, cytokine, genomic, methylation and dietary data from 43 young adult monozygotic twin pairs (aged 22-36 years, 53% female), where 25 of the twin pairs were substantially weight discordant (delta body mass index > 3 kg m-2). These measurements were originally taken as part of the TwinFat study, a substudy of The Finnish Twin Cohort study. These five large multivariate datasets (comprising 42, 71, 1587, 1605 and 63 variables, respectively) were jointly analysed using an integrative machine learning method called group factor analysis (GFA) to offer new hypotheses into the multi-molecular-level interactions associated with the development of obesity. New potential links between cytokines and weight gain are identified, as well as associations between dietary, inflammatory and epigenetic factors. This encouraging case study aims to enthuse the research community to boldly attempt new machine learning approaches which have the potential to yield novel and unintuitive hypotheses. The source code of the GFA method is publically available as the R package GFA.

SUBMITTER: Kibble M

PROVIDER: S-EPMC7657920 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

An integrative machine learning approach to discovering multi-level molecular mechanisms of obesity using data from monozygotic twin pairs.

Kibble Milla M Khan Suleiman A SA Ammad-Ud-Din Muhammad M Bollepalli Sailalitha S Palviainen Teemu T Kaprio Jaakko J Pietiläinen Kirsi H KH Ollikainen Miina M

Royal Society open science 20201021 10

We combined clinical, cytokine, genomic, methylation and dietary data from 43 young adult monozygotic twin pairs (aged 22-36 years, 53% female), where 25 of the twin pairs were substantially weight discordant (delta body mass index > 3 kg m<sup>-2</sup>). These measurements were originally taken as part of the TwinFat study, a substudy of The Finnish Twin Cohort study. These five large multivariate datasets (comprising 42, 71, 1587, 1605 and 63 variables, respectively) were jointly analysed usin ...[more]

PMID: 33204460

Similar Datasets

Project description:Background Subcutaneous adipose tissue (SAT) undergoes major changes in obesity, but little is known about the whole-genome scale patterns of these changes or about their variation between different obesity subgroups. We sought to compare how transcriptomics profiles in SAT differ between monozygotic (MZ) co-twins who are discordant for body mass index (BMI), whether the profiles vary between twin pairs and whether the variation can be linked to clinical characteristics. Methods We analyzed the transcriptomics (Affymetrix U133 Plus 2.0) patterns of SAT in young MZ twin pairs (n =26, intrapair difference in BMI>3 kg/m2, aged 23-36), from 10 birth cohorts of adult Finnish twins. The clinical data included measurements of body composition, insulin resistance, lipids, and adipokines. Results We found 2108 genes differentially expressed (FDR<0.05) in SAT of the BMI-discordant pairs. Pathway analyses of these genes revealed a significant downregulation of mitochondrial oxidative pathways (p<0.05) and upregulation of inflammation pathways (p<0.05). Hierarchical clustering of heavy/lean twin ratios, representing effects of acquired obesity in the transcriptomics data, revealed three sub-groups with different molecular profiles (FDR<0.05). Analyses comparing these sub-groups showed that, in the heavy co-twins, downregulation of the mitochondrial pathways, especially that of branched chain amino acid (BCAA) degradation was more evident in two last clusters while and upregulation of the inflammatory response was most evident in the last, presumably the unhealthiest cluster. High fasting insulin levels and large adipocyte diameter were the predominant clinical characteristic of the heavy co-twins in this cluster (Bonferroni adjusted p<0.05). Conclusions This is the first study in BMI-discordant MZ twin pairs reporting sub-types of obesity based on both SAT gene expression profiles and clinical traits. We conclude that a decrease in mitochondrial BCAA degradation and an increase in inflammation in SAT co-occur and associate with hyperinsulinemia and large adipocyte size in unhealthy obesity.

Dataset Information

An integrative machine learning approach to discovering multi-level molecular mechanisms of obesity using data from monozygotic twin pairs.

Publications

An integrative machine learning approach to discovering multi-level molecular mechanisms of obesity using data from monozygotic twin pairs.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure