Public defence in Information and Computer Science, M.Sc. Maia Malonzo

Title of the thesis: Computational analyses of transcriptome and DNA methylation data

Doctoral student: Maia Malonzo
Opponent: Dr. Reija Autio, Tampere University
Kustos: Prof. Harri Lähdesmäki, Aalto University School of Science, Department of Computer Science

Gene regulation is the process by which genes are either expressed or repressed. Transcriptome analysis determines which genes are regulated, whereas epigenome analysis, e.g. DNA methylation, determines how genes are regulated. In this work, we performed transcriptome and DNA methylation analyses to better understand stem cell regulation and determine biomarkers, genes with important roles and diagnostic value, of two diseases, Alzheimer's disease and asthma. 

To determine the role of stem cell-specific gene POLR3G, we performed transcriptome analysis and identified both coding and non-coding genes that are regulated by POLR3G which potentially mediates the function of POLR3G in stem cell maintenance. We also compared the dynamics of microRNA Let-7 and protein LIN28 in mouse (mESC) and human (hESC) embryonic stem cells and found that the previous assumption that they function in a negative feedback loop is not true in hESC. 

We analyzed DNA methylation from stem cells with chromosomal abnormalities and found that the CAT gene, known to protect cells against DNA damage and oxidative stress, to be both repressed and hypermethylated compared to normal stem cells. DNA methylation aberrations were also analyzed in blood samples of twins discordant for Alzheimer's disease and found the gene ADARB2 to be differentially methylated in both blood and brain samples. DNA methylation analysis of blood samples from children with asthma revealed genes SMAD3, known to be involved in immune response, to be differentially methylated. Overall, our results provide new insight into molecular level mechanisms in the two diseases and provide candidate biomarkers for further validation of their diagnostic potential. 

We developed two statistical tools for analyzing bisulfite sequencing data. LuxRep is a method that utilizes the biochemistry underlying bisulfite sequencing at the level of technical replicates which allows the inclusion of samples with low bisulfite conversion efficiency which are typically excluded as they lead to overestimation of methylation levels. We show that LuxRep improves accuracy of estimates of methylation levels and inference of differential methylation of CpG sites. LuxHMM is a method for determining differentially methylated regions (DMRs) using hidden Markov model to segment the genome and Bayesian regression to infer differential methylation. We showed that LuxHMM performs competitively compared to other published methods for differential methylation.

Key words: DNA methylation, transcriptome, gene regulation, Alzheimer's disease, Asthma, stem cells, bisulfite sequencing, Bayesian, HMM

Thesis available for public display 10 days prior to the defence at: https://aaltodoc.aalto.fi/doc_public/eonly/riiputus/ 

Contact information:

Doctoral theses of the School of Science: https://aaltodoc.aalto.fi/handle/123456789/52

Zoom Quick Guide: https://www.aalto.fi/en/services/zoom-quick-guide