Presenting Author Danbury Math Academy Fairfield , Connecticut, United States
Zehou Zhao (Danbury Math Academy)| Xiaodi Wang (Danbury Math Academy)
Adhesome, the complement of cell-cell and cell-matrix adhesion receptors in an organism, encompasses 150 bona fide components and 82 associated components. The core of adhesome are integrins, a superfamily of cell adhesion receptors, comprising of 24 transmembrane αβ heterodimers via combination of 18 α and 8 β subunits encoded by 26 genes in humans. However, tissue and cell-type specific adhesomes remain elusive. Recent single-cell RNA sequencing (scRNA-seq) data enable characterizing the cell-type specific expression signature of integrins. in this research, we present the weighted bipartite integrin expression maps of 173 cell types during human fetal development and 101 cell types of human intestine. These maps can also be transformed to fuzzy hypergraphs, namely, in which each cell type is represented as a hypergraph edge of fuzzy set of integrin subunits weighted by gene expression levels. A fuzzy hypergraph is predictive for simple cell type identification with single-cell gene expression data through simple logistic regression models. The bipartite networks can be extended to the whole adhesome. Our bipartite network algorithm can also be expanded to multipartite graphs, in future research, with additional layers’ information, such as genome-wide association disease mapping, drug targets, et al. Thus, our algorithm of visualizing and modeling the integrin adhesion code may help us better understand the complexity of molecular underpinning of adhesome thereby fostering adhesome-based molecular medicine.
Bipartite graph expressing cell specific integrin expression with weighted expression over 500 Bipartite Graph expressing cell specific integrin expression