New Methods of Cell Characterization

Computer scientists at Carnegie Mellon University conducted a study that revealed that neural networks and supervised machine learning techniques can efficiently be used to characterize cells that have been studied using single cell RNA-sequencing (scRNA-seq). This research is extremely useful in many ways. It brings about more clarity regarding the new cell subtypes and helps the researchers to chalk out the differences between healthy and diseased cells, or a young versus aged cell.

The new technique analyzes all of the scRNA-seq data and selects only those parameters that can differentiate one cell from another. Earlier the researchers used to employ the single cell sequencing techniques only for the purposes of their research. This new technique is capable of conducting analytical or comparison study for various types of cells. This new technique is also very helpful for the National Institutes of Health's new Human BioMolecular Atlas Program (HuBMAP), which is building a 3-D map of the human body that shows how tissues differ on a cellular level. The experts have also developed an automated pipeline that attempts to download all public scRNA-seq data available for mice—identifying the genes and proteins expressed in each cell—from the largest data repositories, including the NIH's Gene Expression Omnibus (GEO). The cells were then labeled by type and processed with the help of a neural network, a computer system modeled on the human brain. The technique is used to compare all the cells with one another. This way the researchers are able to chalk out the standards for differentiating cells amongst themselves. The model has been tested using a scRNA-seq data from a mouse study of a disease similar to Alzheimer's. The research has revealed that similar levels of brain cells were present in both healthy and diseased cells. The diseased cells consisted of substantially more immune cells just as the macrophages, developed in response to the disease. The researchers employed their pipeline and methods to create scQuery. scQuery, is a web server capable of conducting speedy comparative analysis of new scRNA-seq data. As and when a researcher submits a single cell experiment to the server, the neural networks of the group and matching methods quickly identifies and finds out the cell subtypes and brings forth earlier studies of similar cells. This new method has been described in the journal Nature Communications recently. It shall soon be used as part of the National Institutes of Health’s new Human BioMolecular Atlas Program.

By: Anuja Arora

Content: https://www.news-medical.net/news/20181113/Neural-network-could-replace-marker-genes-in-RNA-sequencing.aspx