Need Help?
Copied to clipboard!

The Finland-United States Investigation of NIDDM Genetics (FUSION) Tissue Biopsy Study

The Finland-United States Investigation of NIDDM Genetics (FUSION) study is a long-term effort to identify genetic variants that predispose to type 2 diabetes (T2D) or that impact the variability of T2D-related quantitative traits (QTs). Most of the variants associated with T2D and related traits (glucose and insulin, anthropometrics, lipids) through genome-wide association studies (GWAS) occur in non-coding regions, suggesting a strong regulatory component to disease susceptibility. Regulatory element activity is often tissue-specific, which further complicates discovery of the causal/functional variation. Therefore, there is a critical need to identify the appropriate cell type, regulatory elements, target genes, and causal variants(s) in T2D-relevant tissues.
We hypothesize that a subset of T2D and related variants alter gene expression regulation in skeletal muscle and adipose tissue - two major insulin target tissues and play key roles in insulin resistance. To that end, our study contains a comprehensive survey of genomics, epigenomics and transcriptomics in skeletal muscle and adipose tissue from individuals with glucose tolerance categories ranging from normal to T2D.

For this FUSION Tissue Biopsy Study, we obtained RNA-Seq, microRNA (miRNA)-Seq, and DNA methylation (methyl)-Seq data on biopsy samples from 331 individuals from across the range of glucose tolerance: 124 normal glucose tolerance (NGT), 77 impaired glucose tolerance (IGT), 44 impaired fasting glucose (IFG), and 86 newly-diagnosed T2Ds. Participants completed two study visits, two weeks apart. First visits comprised most of the clinical phenotyping, including four-point OGTT (fasting, and 30, 60, and 120 minute post-load); BMI, WHR; lipids; blood pressure; and many other variables. Participants also completed FUSION health history, medication, and lifestyle questionnaires. On the second visit, we obtained ~250mg vastus lateralis skeletal muscle, ~750mg abdominal subcutaneous adipose, and a ~5x15mm section of abdominal skin. Visits were completed in March 2013. RNA isolation is ongoing in the Collins laboratory at the NIH, RNA and miRNA sequencing at the NIH Intramural Sequencing Center (NISC), and genotyping at the Center for Inherited Disease Research (CIDR). Individual-level data is available here for the 306 individuals who consented to data deposit. To focus on evaluation of gene expression and its regulation in skeletal muscle, we analyzed mRNA extracted from vastus lateralis skeletal muscle obtained from 271 of the 331 individual subjects from Finland, along with genome-wide genotypes. Individual-level data is available here for the 250 subjects who consented to the use of their data.

Release phs001048.v2.p1 adds muscle data for an additional 42 subjects and data from adipose tissue for 276 subjects. Total RNA was isolated using Trizol extraction in the Collins laboratory at the NIH. The mRNA was poly-A selected, 24-plex libraries were generated using the Illumina TruSeq directional mRNA-seq library protocol and RNA sequencing was performed on HiSeq2000 sequencers using 101bp paired-end reads at NISC. miRNA libraries were prepared from total RNA from 296 muscle and 270 adipose samples, pooled and sequenced 50bp single-end reads on Illumina HiSeq2500. Data for 272 muscle and 251 adipose samples are available here for individuals with consent for data deposit. DNA was extracted from blood in the Collins laboratory, and genotyping on the Illumina Omni2.5M array was performed at CIDR. Genotypes were imputed using the HRC 2016 reference panel. In order to assess regions of open chromatin in skeletal muscle, we obtained muscle tissue from a commercial provider to perform ATAC-seq; these samples were sequenced at the University of Michigan DNA Sequencing Core.

Release phs001048.v3.p1 adds single-nucleus (sn) RNA-seq and ATAC-seq data in 287 skeletal muscle samples out of the original 331 individuals. Individual-level data is available here for the 265 subjects who consented to the use of their data. The frozen tissue biopsy samples were processed in ten batches, each consisting of 40-41 samples. These batches were organized using a randomized block design to protect against experimental contrasts of interest, including cohort, age, sex, and BMI, among others. Samples in each batch were pulverized, pooled together followed by nuclei isolation. The nuclei were processed on the 10X Genomics Chromium platform separately for snATAC-seq and snRNA-seq (v. 3.1 chemistry for snRNA-seq).

Release phs001048.v4.p1 adds additional phenotypes and provides some corrections to several previous phenotypes.

Release phs001048.v5.p1 adds the following:

(1) additional skeletal muscle single-nucleus data: a) snRNA-seq gene-by-count and snATAC-seq peak-by-count matrices, filtered from the phs001048.v3.p1 matrices to include only molecular features with at least one count in 10% of nuclei at a sex- and cell-type-specific level, and b) snATAC-seq nfrPeak count matrices and nucPeak occupancy score matrices.

(2) snRNA-seq data for 189 frozen subcutaneous adipose tissue samples from 184 of the original 331 individuals. Nuclei isolation for 110 of the samples was performed in the Pajukanta lab at UCLA across 8 batches and the libraries were constructed using the 10X Genomics Chromium Single Cell 3' v3.1 platform. These libraries were sequenced at the UCLA Technology Center for Genomics & Bioinformatics on one lane of a NovaSeq 6000 S1 flow cell for one batch and across 6 lanes of a NovaSeq 6000 S4 flow cell for 7 batches. For 79 samples, nuclei isolation was performed in the Mohlke lab at UNC as part of two 64-sample batches, droplet formation was performed at the UNC Advanced Analytics Core on the 10X Genomics Chromium platform, 3' libraries (v3.1 dual index) were generated, and libraries in each batch were sequenced at the New York Genome Center across 4 lanes of a NovaSeq X 25B flow cell.

(3) data from iPSC lines generated from 27 individuals and differentiated to fibro-adipogenic progenitors (FAPs). From a pooled subset of 10 of these individuals, we generated snATAC-seq data at the stem cell, intermediate, and final FAP stages. We performed flow cytometry analysis at the same time points for this pooled subset and two additional individuals. For all 27 individuals, we provide light whole genome sequencing, flow cytometry, and insulin-stimulated glucose uptake assay results. In addition, snRNA-seq and snATAC-seq data were generated for all 27 individuals pooled together under basal and high-insulin conditions. Nuclei were isolated at the University of Michigan, assembled using the 10X Genomics Epi Multiome ATAC + Gene Expression kit, and sequenced at the University of Michigan Sequencing Core on an Illumina HiSeq2500.

In the future, FUSION data will be submitted to the European Genome-phenome Archive (EGA).