genom and cow

Research section: Systems Genomics

The Systems Genomics Section at DTU Bioinformatics conducts research and development in genome-wide association studies, whole genomic prediction of disease risks and performance using data from a large number of individuals, disease-patient stratification using multi-omic data integration and systems genomic analysis, construction of gene- and protein networks, molecular pathway profiling and functional and molecular validations of genes and biomarkers for diseases and quantitative phenotypes.

Genomics involves high-throughput genome-wide measurements of genetic variation among individuals in a population through microarray-based techniques or next or third generation genome sequencing techniques and analysis of such vastly huge high-throughput genomic data to identify causal and regulatory genes and pathways and infer individuals’ disease risks or performance or other phenotypic outcomes.

Genomics is also concerned with the structure, function, comparison, and evolution of genomes. Analysis of high-throughput genomic data requires bioinformatics, statistical or quantitative genomic and computational biology methods to analyze the function and structure of entire genomes and associate genomic architecture of individuals with their disease risks or phenotypic profile.

What is systems genomics?

Systems genomics has genomics as its core but is related to systematic integration and analyses of genomic information or data with other biological data types. Recently, hugely comprehensive multiple omic datasets, often using NGS and other high-throughput technologies, are now available at individual- and population-level where they are characterized or observed for diseases and other complex quantitative traits.

Data types come from genome-wide, epigenome-wide, transcriptome-wide, proteome-wide, metabolomic and metagenomic measurements. On top of these are lifestyle and environmental data.

Systems genomics integrates, models and analyzes these multi-omic big datasets with an aim to link biological variations at the individual’s and at the population’s level with observed traits or clinical diseases and environmental life style characteristics. Inherent in the work is deep understanding and profiling of biological pathways and biological networks that underlie the disease or phenotype manifestations.

The “systems” aspects of systems genomics is therefore focused on testing and correlating genetic variants for a range of intermediate molecular or endo-phenotypes and observed phenotypes in individuals as well as characterizing those parts of the molecular networks that drive these complex phenotypes.

pc and data

Using big data and ancestry to predict disease risks and complex quantitative traits or phenotypes

In our section, we use genomic data together with well-characterized phenotypes or diseases and ancestral and pedigree data to predict genetic risks for diseases, using quantitative genetics / statistical genetics methods. The genomic prediction includes both array-based genotype data and whole genome DNA sequence data using NGS technologies.

For genomic prediction, we use various methods including standard multiple regression models through complex BLUP mixed modes, Bayesian methods, machine learning to Artificial Intelligence (AI) methods.

We also develop data integration methods, statistical models and bioinformatics and computational strategies to analyze multi-omic datasets from individuals or populations. By doing this, we study genetic variations of intermediate molecular phenotypes and complex biological networks that eventuate as an observed trait or clinical diseases at the animal or human level.

For a recent review article on this see our paper in Genetics, Selection and Evolution (2016). We apply systems genetics approaches for detecting causal genetic factors, predictive biomarkers and constructing causal and regulatory networks underlying important diseases and traits.

Towards precision and personalized medicine

Personalized medicine is an approach to the practice of medicine that uses information about a patient’s unique proteogenomic makeup, life style attributes and environmental conditions to provide customized or individualized treatments to diseases, rather than “one-size-fits-all” approach.

Currently with the help of a wealth of clinical, NGS-based genomic and other high-throughput assay data, we can design tailored therapies for patients. Denmark in particular is well suited as we have centralized registry information of patients – a unique strength of the Danish medical system.

We have strong collaborations with major hospitals in Denmark to develop better strategies for the implementation of immunotherapies against cancer. For these projects we also rely on intensive genomic profiling of individual tumors, while at the same time performing deep immune profiling to systemically map tumor-immune system interactions.

We have ongoing projects on checkpoint inhibitor therapy of glioblastoma multiforme, adoptive cell transfer therapy for melanoma, and a large-scale proteogenomic profiling of breast cancer for precision application of existing combination therapies. To ensure not only clinical application, but also basic scientific knowledge output, we have collaborative partners at top academic researchers at University of Oxford, Stanford University, and Heidelberg University.

Read the DTU web news Better immunotherapy for cancer patients.

Highlights of research

Some highlights of our research that enabled precision and personalized medicine in Denmark can be seen from the recent outcome of Genome Denmark project - the Danish reference genome published in the prestigious journal Nature (2017). Read more in the DTU web news Danish reference genome now mapped.

This high-depth, high-coverage de novo assembled genome ranks best among similar population-specific assembled reference genomes and constitutes an unprecedented resource for the identification of causative mutations in complex human disease. By helping to separate the wheat from the chaff (i.e. causative vs neutral variants), the Danish reference genome positions Danish taxpayers at the forefront of personalized healthcare and treatment.

Another example is that we have also linked certain genetic factors that influence early child growth to adult diseases via shared patho-physiological mechanisms – a result of massive international collaboration. By screening children at a young age, we can via their DNA profiles foresee whether they are at risk of developing lifestyle diseases such as type 2 diabetes and cardio-metabolic problems.

If so, we can establish prevention work in terms of diet and exercise and offer better treatments suited to their genetic profiles so called “precision medicine”. The study has been published in the article "Genome-wide associations for birth weight and correlations with adult disease" in the prestigious journal Nature (2016).

Research approaches

Our section works on combining genomic and molecular level systems biology data with such health informatics data from the healthcare sector, such as for example electronic patient records and biobank questionnaires. Similar approaches are taken for predicting and improving animal health and performance using veterinary clinical data.

We also develop research strategies relying on integration of massive amounts of experimental or field data for conventional systems levels analysis, but also in the context of precision medicine and one health (animal and human). The aim is to combine and stratify patients not only from their genotypes, but also phenotypically based on the clinical descriptions in the medical records which describe disease development in detail.

Research Projects in the section of systems genomics is focused on unravelling the molecular causes of complex human and animal diseases including obesity, diabetes, congenital cardiopathies, cancer and animal diseases. The section is actively involved in many key consortia driving research in these areas.

DTU Multi Assay Core Facility (DMAC)

DMAC provides Next and Third generation sequencing service for Transcriptome-, Whole genome-, Amplicon-, Exome and custom designed capture-, Chip-seq- and metagenomics- Sequencing. Services include sample preparation and qualitative and quantitative assessment of RNA/DNA samples for various applications and sequencing, which can be either performed in-house or outsourced as a service with data transfer.

DMAC is hosted within the Section of Systems Genomics while it is a core facility for the entire department and DTU and is a service provider for hospitals and other university groups. 
Read more about DMAC.

Supercomputing solutions for big omics data

Our department and hence section of systems genomics has unique access to and a significant share of storage and computing resources of the national supercomputer for life sciences, called the “Computerome”. It is an important backbone for most of our research at DTU Bioinformatics. It is physically installed at DTU Risø Campus.

With this platform, we can collaborate with human and veterinary medicine science related partners who generate big data and have supercomputing needs to store and analyze data in secure environment.

Impact to society and science

We work to make an impact in the agri-biotech and biomedical industries and we aim for being internationally recognized as pioneers in the integration of multi-omics big data, quantitative genomics and data-driven systems biology research, including supercomputing.

With this, we are well positioned to significantly contribute to national initiatives such as: personalized, precision and predictive medicine, One Health - human and animal health and supercomputing solutions for big omics data.


Haja Kadarmideen
Head of Department, Professor
DTU Bioinformatics
+45 45 25 61 61
22 FEBRUARY 2018