Using Various Types of NGS Data for Understanding Cell Type-Specific Regulation of Gene Expression
By Kenta Nakai, The Institute of Medical Science, the University of Tokyo, Tokyo, Japan

Abstract:
Although about 20 years have passed since the completion of the Human Genome Project, understanding how the genome DNA sequence orchestrates life activities is still premature. To address this kind of issues, a knowledge-based system named PSORT, was constructed which predicts the subcellular localization sites. However, with the progress on our understanding of the protein localization mechanisms as well as the growth of proteins with known subcellular localization, promptly updating and optimizing the rule-based system became difficult. Thus, a new version of the prediction system, PSORT II, based on a machine learning algorithm (the k-nearest neighbor method) was constructed. Similar approach could be useful in characterizing the regulatory regions of genes that direct their tissue-specific expression. With this approach, genes that are specifically expressed in 26 human and 34 mouse tissues were analysed. Another similar approach to characterize the cis-regulatory architecture of genes in relatively smaller genomes, such as those of Arabidopsis and Drosophila was attempted and extended it to Caenorhabditis elegans genes, based on scRNA-seq data. The aim was to expand the number of genes that harbor characteristic cis regulatory architectures, leading to the construction of an encyclopedia of gene regulatory regions in C. elegans. In the talk, some of its preliminary results are discussed. In addition, it seems likely that a mere combination of transcription factor binding sites is not enough to explain the entire cell type-specific gene expression. Although their detailed mechanisms are still mysterious, the changes of the higher-order chromatin structures are also responsible for the cell type-specific expression of genes and such changes are related to the changes of epigenetic marks, such as the modification pattern of histones and DNA methylation pattern. Interestingly, the information of such epigenetic patterns can be obtained from NGS (next generation sequencing) data, such as ChIP-seq and WGBS; the information of 3D chromatin structure changes is also obtainable from Hi-C or its relative experiments. Some of the studies based on these kinds of data are introduced here.

This keynote is a part of InSyB2020 ([ Ссылка ]) which is organized by APBioNET ([ Ссылка ]).

свернуть