Организация, в которой проходила защита:
Институт биоинформатики (Санкт-Петербург)
Год защиты:2019
Аннотация:Environmental factors, including chemicals, can cause epigenetic changes that can be traced to subsequent generations. The most studied epigenetic changes are DNA methylation, small non-coding RNA, and histone modification. Smoking remains one of the most adverse voluntary health risks. Reduced representation bisulfite sequencing (RRBS) data can be used to study the pattern of methylation changes upon exposure to smoking.
Our project is a part of the Russian Children’s Study, a prospective cohort of 516 boys who were enrolled at 8–9 years of age and provided semen and blood samples at 18–19 years of age (Sergeyev et al, 2017). We analysed smoking influence on the DNA methylation level of peripheral blood leukocytes at the age of 18. To search for differentially methylated CpG islands and regions (DMR), we used two different approaches.
To implement the first approach, we used data from the CpG islands presented in all samples of peripheral blood. After exclusion regions that did not meet the inclusion criteria, we used the A-clustering algorithm (Sofer et al, 2013) to combine the regions into clusters and generalized estimating equation model to search for significant DMRs. With this approach we identified 217 A-clusters, from them 77 were significant (p-value < 0.05). Because for A-clustering implementation we needed to restrict our data, a lot of important information could have been lost.
DMRcate - R package for search of differentially methylated regions (DMRs) associated with exposure to a factor (Peters at al, 2015). In our project, we used smoking past half year in binary classification (smoke or not) to find DMRs associated with exposure. 145 significant CpGs and 34 significant DMRs (p-value < 0.05) were found in our data. From them 19 DMRs overlap with at least one promoter (reference - GRCh 38). We found 23 genes associated with significant DMRs. These genes are associated with antisense RNAs, lincRNAs, miRNA, pseudogenes, zinc fingers, transcriptional factor, spliceosome, cell adhesion and migration, kinase, metalloprotease, electron transport chain and amino-acid transporter.
Finally, we found DMRs using two different statistical strategies for analysis of DNA RRBS data. Further research plans include the analysis of changes in the expression of various groups of small RNAs, as well as a comparative analysis of leukocyte and semen RRBS data. The project is available in our github repository: https://github.com/mariafiruleva/leykocytes_dmr_analysis.