Statistical and computational analysis of high-throughput ‘omics’ datasets for understanding the etiology and pathogenesis of autoimmune diseases

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
School of Science | Doctoral thesis (article-based) | Defence date: 2021-08-16
Degree programme
150 + app. 164
Aalto University publication series DOCTORAL DISSERTATIONS, 90/2021
The primary function of the human immune system is to maintain our wellbeing by protecting ourselves from harmful substances and microbes (i.e. pathogens) that we continuously encounter through our surroundings. However, a variety of factors can lead to immune system dysfunction, which in turn can give rise to various incurable diseases, including autoimmune diseases (ADs), such as type 1 diabetes (T1D), immunoglobulin G4 related disease (IgG4-RD) and systemic sclerosis (SSc). In ADs, the immune system fails to distinguish between pathogens and body's own cells, and mistakenly attacks body's healthy tissues. Unfortunately, the factors that trigger ADs (i.e. etiology) and the molecular mechanisms by which ADs develop (i.e. pathogenesis) remain poorly understood. Genetics and environmental factors, such as gut microbiome, have been implicated in triggering or influencing the development of ADs, but the concerned mechanisms remain largely elusive. Therefore, the aim of this thesis is to further our understanding about the etiology and pathogenesis of ADs by performing robust statistical and computational analyses on high-throughput 'omics' datasets. More specifically, one of the aims of this thesis was to study transcriptomics data from immune cells of T1D susceptible infants in order to identify gene expression markers that can aid in predicting the onset of autoimmunity and/or characterizing the disease progression. We found several genes to be associated with the pathogenesis of T1D, including IL32 that has not been associated with T1D before. Another aim was to develop a personalised method that can robustly model longitudinal transcriptomics data from heterogeneous diseases and identify pathways associated with the pathogenesis of the disease. When applied to T1D data, this method was able to associate several key pathways to T1D pathogenesis that were missed by other methods. Additionally, this thesis aimed to study the gut microbial architecture of IgG4-RD and SSc patients (metagenomics data) and identify potential sources of microbial signals that may be contributing to the etiology of the two diseases. Among other interesting results, we found a specific strain of Eggerthella lenta that contains genes with the potential of influencing the immune system, to be significantly overabundant in both diseases. Finally, this thesis also aimed to identify the environmental and host-related factors that may be influencing the development of the highly dynamic early gut microbiome of T1D susceptible infants. In effect, we linked several new factors to the development of the early gut, such as household location at birth, maternal antibiotic treatments and average increase in height and weight per year, to name a few.
Defence is held on 16.8.2021 12:00 – 16:00 via remote technology (Zoom),
Supervising professor
Lähdesmäki, Harri, Prof., Aalto University, Department of Computer Science, Finland
Thesis advisor
Lähdesmäki, Harri, Prof., Aalto University, Department of Computer Science, Finland
high-throughput data, statistical modelling, computational analysis, immune system, gut microbiome, genes, pathways, autoimmune diseases
Other note
  • [Publication 1]: Henna Kallionpää, Juhi Somani, Soile Tuomela, Ubaid Ullah, Rafael de Albuquerque, Tapio Lönnberg, Elina Komsi, Heli Siljander, Jarno Honkanen, Taina Härkönen, Aleksandr Peet, Vallo Tillmann, Vikash Chandra, Mahesh Kumar Anagandula, Gun Frisk,Timo Otonkoski, Omid Rasool, Riikka Lund, Harri Lähdesmäki, Mikael Knip, and Riitta Lahesmaa. Early Detection of Peripheral Blood Cell Signature in Children Developing β-Cell Autoimmunity at a Young Age. Diabetes, vol. 68, pp. 2024-2034, October 2019.
    Full text in Acris/Aaltodoc:
    DOI: 10.2337/db19-0287 View at publisher
  • [Publication 2]: Juhi Somani, Siddharth Ramchandran, Harri Lähdesmäki. A personalised approach for identifying disease-relevant pathways in heterogeneous diseases. npj Systems Biology and Applications, vol. 6, article number 17, June 2020.
    Full text in Acris/Aaltodoc:
    DOI: 10.1038/s41540-020-0130-3 View at publisher
  • [Publication 3]: Damian R. Plichta, Juhi Somani, Matthieu Pichaud, Zachary S. Wallace, Ana D. Fernandes, Cory A. Perugino, Harri Lähdesmäki, John H. Stone, Hera Vlamakis, Daniel C. Chung, Dinesh Khanna, Shiv Pillai, Ramnik J. Xavier. Congruent microbiome signatures in fibrosis-prone autoimmune diseases: IgG4-related disease and systemic sclerosis. Genome Medicine, vol. 13, article no. 35, February 2021.
    Full text in Acris/Aaltodoc:
    DOI: 10.1186/s13073-021-00853-7 View at publisher
  • [Publication 4]: Tommi Vatanen, Damian R. Plichta, Juhi Somani, Philipp C. Münch, Timothy D. Arthur, Andrew Brantley Hall, Sabine Rudolf, Edward J. Oakeley, Xiaobo Ke, Rachek A. Young, Henry J. Haiser, Raivo Kolde, Moran Yassour, Kristiina Luopajärvi, Heli Siljander, Suvi M. Virtanen, Jorma Ilonen, Raivo Uibo, Vallo Tillmann, Sergei Mokurov, Natalya Dorshakova, Jeffrey A. Porter, Alice C. McHardy, Harri Lähdesmäki, Hera Vlamakis, Curtis Huttenhower, Mikeal Knip, and Ramnik J. Xavier. Genomic variation and strain-specific functional adaptation in the human gut microbiome during early life. Nature Microbiology, vol. 4, pp. 470-479, March 2019.
    Full text in Acris/Aaltodoc:
    DOI: 10.1038/s41564-018-0321-5 View at publisher