Nature Biotechnology: A new tool to understand how the genome is organized in a single cell

Within the microscopic boundaries of a single human cell, the complex folding and arrangement of protein and DNA bundles determine the fate of a person: which genes are expressed, which genes are suppressed, and-importantly-whether they stay healthy or suffer Disease

Although these gene bundles have potential effects on human health, scientists know little about how the genome folds in the nucleus and how it affects gene expression

The algorithm is called Higashi and is based on hypergraph representation learning-a form of machine learning that can recommend music in applications and perform 3D object recognition

PhD student Ruochi Zhang and PhD student Tianming Zhou in the School of Computer Science and Jian Ma, Professor of Computational Biology Ray and Stephanie Lane, led the project

"He goes to research with enthusiasm, but sometimes with a sense of humor," Professor Ma said

Their research was published in the journal Nature Biotechnology as part of a multi-institutional research center that seeks to better understand the three-dimensional structure of the cell nucleus and how this structural change affects cell function in health and disease

This algorithm is the first tool to use complex neural networks on hypermaps to provide high-resolution analysis of single-cell genome organization

Chromosomes are composed of a dna-rna-protein complex called chromatin, which folds and arranges itself to fit the nucleus

The Higashi algorithm works with an emerging technology called single-cell Hi-C, which creates a snapshot of simultaneous chromatin interactions in a single cell

"The variability of genome organization has a strong suggestive effect on gene expression and cell state

The Higashi algorithm also allows scientists to simultaneously analyze other genomic signals that are analyzed together with single-cell Hi-C

"This is a rapidly developing field," Professor Ma said

Original title:

Multiscale and integrative single-cell Hi-C analysis with Higashi