RNA Sequencing in the Framingham Heart Study Third Generation Cohort Exam 2

RNA sequencing (RNA-seq) is a powerful tool to evaluate the transcriptome with incredible depth and clarity. As compared to gene expression arrays, RNA-seq allows the identification and quantification of a larger set of known transcripts (including long non-coding RNAs [lncRNAs]), novel transcripts, alternative splicing events, and allele-specific expression (including parent-of-origin allele-specific expression); all with a vastly higher signal-to-noise ratio compared to gene expression profiling via microarrays. The relations of these transcriptomic features to health and disease in very large population studies is underexplored. It is our belief that this proposed project will identify new biomarkers of disease risk and provide insights into disease pathogenesis. The Framingham Heart Study (FHS) is uniquely suited to conduct RNA-seq because of the wealth of existing phenotype resources in conjunction with whole genome sequence (WGS) data from TOPMed and methylomic data, data and other omics data that can be leveraged at extremely low cost to maximize the impact of an investment in RNA-seq.

The advent of high-throughput RNA-seq technology has revolutionized transcriptomic profiling at an unprecedented scale, leading to the discovery of new RNA species and deepening our understanding of transcriptomic dynamics. Compared to microarray-based RNA profiling, RNA-seq is appreciated for its ability to reveal the complexity of the transcriptome, encompassing previously unknown coding and lncRNA species, novel transcribed regions, alternative splicing, allele-specific expression, and fusion genes This project proposes to build upon and extend the work conducted using gene expression arrays in the FHS by examining complex transcriptomic features that cannot be determined using microarray-based expression data.

In this proposal we focus on expression levels of protein-coding RNAs, lncRNAs, alternative splicing, and allele-specific expression. There are \~18,000 mRNA transcripts at the gene-level for protein-coding RNAs. Alternative splicing is a tightly regulated process that produces different mRNA isoforms from genes that contain multiple exons. One major application of RNA-seq is to detect even subtle differences in exon splicing. lncRNAs are non-protein coding transcripts longer than 200 nucleotides and have been implicated in many biological process. For example, some lncRNAs impact the expression of nearby protein-coding genes, some can bind to enzymes regulating transcription patterns, and other lncRNAs are precursors of small RNAs. A number of computational methods have been developed to detect alternative splicing and lncRNAs from RNA-seq data. Identification of alternative splicing and lncRNAs will be standardized across TOPMed studies and we will conduct analyses on centrally called splice data as well as lncRNAs. Allele-specific expression (ASE), which cannot be measured using microarrays, allows the differentiation between transcripts from the two haplotypes of an individual at heterozygous sites. ASE enables a more granular understanding of how a disease-related genotype affects gene expression. ASE has been linked to human disease in small sample sets but has not been examined fully in large populations. Standard

bioinformatics tools have been developed to study ASE. In addition, with TOPMed WGS data on parents from the FHS Offspring cohort, it will be possible to study parent-of-origin ASE, thus furthering our ability to dissect factors that contribute to the transgenerational inheritance of cardiometabolic disease.

In this Application, we propose to extend the investigation of transcriptomics in FHS Third Generation cohort exam 2 participants. The aims of conducting RNA-seq in the FHS Third Generation cohort mirror and extend those of our original microarray-based gene expression profiling. Specifically, we will examine the association of complex transcriptomic variation to: 1) cardiometabolic disease outcomes, 2) genetic sequence variation, and 3) multiple layers of omic data (Aims 1-3). With the proposed RNA-seq data, investigators as well as the general scientific community (via dbGaP access) will have the ability to study transcriptomics from different perspectives always leveraging existing resources to advance the scientific value of this project. To maximize the return on investment, sequencing will be performed by a designated TOPMed RNA-seq laboratory, and the aims of this project will be coordinated with other

TOPMed studies that are conducting RNA-seq.