Characterization of population specific variation in non-coding genomic regions and application to Type 2 diabetes

Type 2 Diabetes Mellitus (T2DM) is a metabolic disorder characterized by reduction in insulin sensitivity. It is a global problem (more than 6% of the world’s population are estimated to be living with T2DM) reducing quality of life and representing a major socioeconomic burden. T2DM incidence varies significantly by region but genetic studies are predominantly in European populations. Also, these studies are reaching a limit in terms of identifying diagnostic value. We have performed a multi-population study of T2DM to investigate the role of the non-coding genome, the regions that don’t code for genes. We have investigated how how T2DM disrupts gene regulation and how this chanbge can be used to profile T2DM risk and disease progression. Genome wide association studies (GWAS) are commonly used to identify associations between single nucleotide polymorphisms (SNPs) occurring within exons, the regions that code for proteins, and disease (e.g. coronary artery disease) or trait (e.g. high blood pressure). A typical GWAS study would investigate two study groups, for example, normal blood pressure and high blood pressure, and attempt to identify SNPs that are specific to one group and which are located within or nearby proteins that have known association with hypertension. While GWAS have identified associations between SNPs and disease it has only for been a relatively small number of SNPs, and new approaches are needed to increase the ‘diagnostic yield’ of these studies. In this work, we looked for associations between SNPs located outside the exons and T2DM. However, the non-coding regions represent ~98% of the genome, so we have focused on specific regions that are associated with microRNAs (miRNAs), short pieces of RNA (typically 22 nucleotides) that regulate gene expression. While there has been an increase in sequencing of populations from low- and middle-income countries, both GWAS and T2DM studies remain focused on high income countries. Such an imbalance can result in inaccurate risk assessment in understudied populations, and ineffective clinical practice and public health policies. There are efforts to address this problem (e.g., the 1000 Genomes Project, H3Africa and the 100K Han Chinese project) but analyses and interpretation of these data is far from complete. In this work, we have been reexamining data from these projects and performing our own sequencing studies of underrepresented populations to perform the first multi-population study of T2DM. In addition to providing new insight into T2DM disease mechanisms and risk, our project yields invaluable insight into overcoming the challenges of performing multi-population studies to motivate studies of other diseases. miRNAs play an important role in gene regulation and the presence of a SNP in their sequence can alter (i) how they effectively regulate their gene targets and (ii) which genes they regulate. We have performed a comprehensive study of how different SNPs are present in different miRNAs in different populations, and how this affects predisposition to T2DM. Our starting point was the 1000 Genome Project, which contains Whole Genome Sequence data for 26 populations from around the world. To investigate this data, we developed our own software tools to perform genome wide characterisation of the regulation networks formed by miRNAs and their target genes. We then investigated how the shape of these networks changed in different populations, and how this affects risk of developing T2DM. In addition, we have collected samples from T2DM Patients and matched Controls from Ethiopia and performed Whole Genome Sequencing, miRNA-seq, RNA-seq and Whole Genome Bisulfide Sequencing. This is the first comprehensive study of its kind and will allow us to investigate how the the networks change between healthy and T2DM patients. We are also in the process of collecting samples from T2DM and matched Controls from The Republic of Congo. By comparing these two study groups, we will be able to compare how the risk changes between these two populations.

We are still in the process of analysing the data we have collected during the project period. However, the data is unique insofar as we are characterising a patient cohort in terms of small RNA seq, mRNA seq, Whole Genome Bisulfite Seq and whole genome seq data. This will allow us to characterise risk in a patient using standard approaches such as identificaiton of SNPs associated with risk. We can also use standard approaches such as looking for statistically significant changes in transcription that are associated with the onset of Type 2 Diabetes. However, what we can also do is use the integrated data to look for significant changes in the structure of the gene regulation networks. this would be a completely different way to evaluate risk and pre-disposition to Type 2 Diabetes. We are still in the process of analysing this data

The project combines computational biology, data mining, genetics, molecular biology and translational research. It will develop novel analytical methods to identify population specific variations in non-coding (NC) regions of the genome, and then apply them to the study of Type 2 Diabetes Mellitus (T2DM). Results will be verified and further investigated experimentally. The novelty of the work is that it hypotheses that, rather than coding regions, the NC regions of the genome that code for the miRNA biogenesis pathway can be used as a starting point for investigating population specific genetic association with disease. The work will lead to a deeper understanding of the miRNAome in the context of disease, and addresses both the issue of significant bias in population in genetic studies, as well as the rising significance of non-communicable diseases such as T2DM1. A key part of the project design is the creation of a multidisciplinary international consortium with the necessary experience and resources to support the proposal, with core expertise (methodology development, genetics & sequencing) located at the host department of Medical Genetics (AMG) at Oslo University Hospital (OUS). Our partners include the Human Longevity Institute (HLI), The African Genome Variation Project (AGVP) and the T2D-GENES and GoT2D consortiums (T2D-Genes/GoT2D). Our partners have published in high profile journals