Application of Data Dimension Reduction Method in High-dimensional Data based on Single-cell 3D Genomic Contact Data

Abstract

The volume and dimensions of data in a variety of fields, especially in biology, are increasing day by day, but our existing analytical methods are difficult to directly apply to high-dimensional data such as single-cell Hi-C Data. Here we perform unsupervised method PCA, t-SNE to reduce the dimensions for data visualization. And we further evaluate the information retention of decomposed components by using LDA classifier model. Our results suggest that those methods can capture and present information that we cannot directly observe.