A team of investigators at the National Centre for Genomic Analysis (CNAG-CRG) in Barcelona has developed a new computational tool, based on the mathematical Graph theory, to infer global, large-scale regulatory networks, from healthy and pathological organs, such as those affected by diabetes or Alzheimer’s disease. Findings from the new study—published recently in Genome Biology through an article titled “Single-cell transcriptomics unveils gene regulatory network plasticity”—were able to pinpoint genes relevant to organ function and potential drivers of diseases.
Although the knowledge we have about human cells and tissues has steadily increased over recent decades, many things remain unknown. For instance, cells exist in transient, dynamic states and understanding them is fundamental to decipher diseases and find cures. Classic techniques used in the lab to study cell types faced limitations and did not enable a finely detailed profile of cell function. To overcome this obstacle, the CNAG-CRG scientists developed this new tool based on their earlier work.
“Our previously developed single-cell transcriptomic tools were very useful to discover unknown cell types,” explained lead study investigator Giovanni Iacono, PhD, senior postdoc researcher at the CNAG-CRG. “Those tools allowed us to describe new types and subtypes of cells, with their unique biological roles and hierarchical relationships.”
Up to now, single-cell analysis had been used to understand cell types and their function within tissues. “Large-scale consortia like the Human Cell Atlas Project generate single-cell maps of entire organisms and sophisticated analysis strategies are required to transform big data into disruptive biological and clinical insights,” stated senior study investigator Holger Heyn, PhD, a team leader of the single cell genomics group at the CNAG-CRG.
The tool that this scientific team has now developed will enable them to go one step further, to see how genes interact to form tissues. “Our tool tries to address precisely the regulatory process that controls the morphology and functions of a cell,” highlighted Iacono.
The new tool is based on Graph theory, an abstract mathematical model in which there are nodes connected by edges. Once you have a graph, a structure, you can measure the importance of each node for the network. In this case, each node was a gene and importance was defined as the function of that gene is key for the biological system under study.
The research team processed datasets from ten-thousands of cells to infer the regulatory networks that drive cell phenotype formation and their respective functions. They applied their tool to study type 2 diabetes and Alzheimer’s disease and were able to find the functional changes relevant to those diseases. Importantly, this opens the door to finding new drug targets.
“The network analysis we have developed goes beyond currently applied approaches to provide deep insights into how gene activities shape tissues and organs. This is critical to understand diseases in which these networks are disrupted and find their ‘Achilles heels’ for effective treatments,” Heyn noted.
Potentially, the tool can be applied to any disease, from Alzheimer’s to chronic lymphocytic leukemia. “We will apply our tool to propose new target genes for many diseases that can then be validated in further studies,” Iacono concluded.