Åsa Johansson – The role of genetic and lifestyle factors on human health and on the risk of common diseases
Every fourth individual in Europe has ever been diagnosed with a common disease, such as cancer, asthma, diabetes, or myocardial infarction. The aim of our research is to determine genetic and lifestyle factors that influence human phenotypes, clinical variables and risk of disease. We also evaluate how such factors can be used to predict, or even prevent, common diseases.
Our goal is to broaden our knowledge of disease pathophysiology, which is important for developing drugs to prevent, delay progression, or relieve symptoms of disease. We have three main research interests:
1. Genetic determinants of complex phenotypes
One of the main focuses in our research has been the genetic contribution to common diseases. During the last 15 years, we have contributed to the identification of thousands of genetic variants that influence the risk of common diseases, using genome-wide association studies (GWAS). Despite this success, only a fraction of the genetic contribution to disease has been identified, and most of the genetic factors remain unknown. Our ongoing projects in genetic epidemiology/genomics medicine focus on:
- Gene-based tests to identify the phenotypic effects of rare genetic variants from whole-genome sequencing data.
- Using Machine Learning / Artificial Intelligence tools for capturing the phenotypic effect of genetic variants.
- Polygenic risk prediction to identify high-risk individuals that are suited for preventive treatments.
- Identifying polymorphic structural variations, that are associated with phenotypic effects, using both short- and long-read sequencing.
2. Epidemiology and biostatistics
Besides genomics, our research also focuses on lifestyle factors and their effects on disease risk. Here we use biostatistical approaches aiming to identify risk factors for common diseases, and to identify gene-environment interactions and sex-specific effects. Some ongoing projects include:
- The effect of obesity-related traits on risk of cancer.
- Using imaging data to fine-map obesity phenotypes to enable precision metabolism.
- The effect of endogenous and exogenous hormones (hormonal contraceptives and hormone replacement therapy) on disease risk with special focus on stroke, depression, bone health, and rheumatic disorders, as well as ovarian, endometrial and breast cancer. Using multi-omics data for characterising disease pathways, and for disease detection and prognosis.
3. Mendelian randomisation and causal inference
One limitation in observational studies is the difficulty to determine causality since observational studies are influenced by confounding and reverse causation. In our research we use several approaches for causal inference, including Mendelian Randomisation which is an instrumental variable approach to obtain consistent estimates of putative causal relationships. Some of our research projects include:
- Using Mendelian randomisation to evaluate the causal effect of inflammatory protein biomarkers on risk of disease. Biomarkers have been identified for many diseases, and might serve as potential drug targets.
- Using non-linear Mendelian randomization to establish causal links between between life style factors and disease risk. Characterising disease pathways using multi-variable Mendelian randomisation.