Quantitative Biology > Molecular Networks
[Submitted on 15 Dec 2023]
Title:Multiscale differential geometry learning of networks with applications to single-cell RNA sequencing data
View PDF HTML (experimental)Abstract:Single-cell RNA sequencing (scRNA-seq) has emerged as a transformative technology, offering unparalleled insights into the intricate landscape of cellular diversity and gene expression dynamics. The analysis of scRNA-seq data poses challenges attributed to both sparsity and the extensive number of genes implicated. An increasing number of computational tools are devised for analyzing and interpreting scRNA-seq data. We present a multiscale differential geometry (MDG) strategy to exploit the geometric and biological properties inherent in scRNA-seq data. We assume that those intrinsic properties of cells lies on a family of low-dimensional manifolds embedded in the high-dimensional space of scRNA-seq data. Subsequently, we explore these properties via multiscale cell-cell interactive manifolds. Our multiscale curvature-based representation serves as a powerful approach to effectively encapsulate the complex relationships in the cell-cell network. We showcase the utility of our novel approach by demonstrating its effectiveness in classifying cell types. This innovative application of differential geometry in scRNA-seq analysis opens new avenues for understanding the intricacies of biological networks and holds great potential for network analysis in other fields.
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.