FiRE Algorithm To Detect Rare Cells Developed By Indian Researchers
FiRE Algorithm To Detect Rare Cells Developed By Indian Researchers
Identifying a rare cell form dataset containing millions of cells is a tedious process. To make it easy, researchers from Delhi have developed a new algorithm that can detect rare cells from a large pool of cells within a second. The finder of rare institutions (FiRE) is an algorithm that is assigned rarity score for each cell based on the gene expression profile of approximately twenty thousand genes. Cells having scores above a certain limit are rare cells. In fact, besides being fast and accurate, it also shows better sensitivity and exclusiveness than current methods.
Some examples of rare cells are transmitted to tumor cells, cancer stem cells, antigen-specific T cells. Identification of these rare cells can open doors to understanding the complex mechanism of these cells, which cause fatal illnesses. This will allow early diagnosis and diagnosis of cancer-like illness.
Professor of Indian Institute of Technology (IIT) Delhi Prof. Jaydev and Indraprastha Institute of Information Technology (IIIT-Delhi)
The team discovered a new sub-type parsed tuberalis cell lineage that is related to the development of pituitary gland.
Prashant Gupta, one of the first authors of the paper from IIT Delhi, said that existing algorithms use clustering or other statistical techniques which contain strict parameter estimates, thus computational cost is very high. He further said that to complete the level of rarity of each cell, FiRE uses the sketching method, which is a type of locality-sensitive hashing. In the hashing technique, cells with similar properties are put together.
Pro. Jaidev said with a background in machine learning that existing equipment is very complex and tedious to detect rare cells when it comes to analyzing a large number of cells. Rare cells can be discovered on large-scale single-cell messenger RNA datasets using FiRe. Gene expression was used as a base to develop an algorithm. Drop-sec further allows effective reading of gene expression profiles of thousands of cells in a short period.
Five different data sets were used to test the viability of the algorithm. The gene expression of about 68,000 different cells was compared to peripheral blood with 0.3% of megakaryocytes, rare cells appeared in rare rare rare cells. As a result, rare cell culture included only megakaryocytes, thus algorithms were validated.
In a simulation experiment to evaluate the performance of the FiRE algorithm, the gene expression profile of two types of cells was mixed in vitro. And by increasing the percentage of a cell type (from 0.5 to 5%), the team tested the accuracy and sensitivity of FiRE and other existing algorithms to correctly identify rare cells. The sensitivity of the FiRE algorithm was still higher than the rest when rare cells contained 0.5% of the population. "When he constituted 2.5%, Fair could identify rare cells with 85% accuracy, which is much more than other algorithms," says Jindal of the Delhi University and second first author of the paper.
"We are now searching for the new cell type [Pars tubelis], which is used by using the FiRE. Most lethal cancerous tumors release the tumor cells. Therefore, we are trying to use FiRE to detect initial cancer by identifying circulating tumor cells, which are rare in peripheral blood,
FiRE Algorithm To Detect Rare Cells Developed By Indian Researchers
Reviewed by Tech Gyan
on
January 07, 2019
Rating:
Reviewed by Tech Gyan
on
January 07, 2019
Rating:
