At the Rapaport Faculty of Medicine, an algorithm is being developed that improves the ability to compare cancer tumors of different patients in a way that overcomes the many differences between them
Researchers at the Rapaport Faculty of Medicine have developed an innovative algorithm that reveals a common and continuous denominator from multidimensional information collected from tumors of different patients. the research published in the journal Cell Systems It was led by Prof. Shay Shen-Or, Dr. Yishai Ofran and Dr. Eilat Alpert from the Rapaport Faculty of Medicine, who conducted it in collaboration between researchers at the Technion, Rambam Medical College, Shaare Zedek and the University of Texas.
Cancer research has undergone significant revolutions in recent years, including the possibility of characterizing a single cell at high resolution, and more specifically - measuring at the same time a large number of genes or proteins from individual cells. This revolution has led to the generation of vast amounts of multidimensional information about large numbers of cells, allowing Determine the characteristics of the healthy tissue and the cancerous tissue. These amounts of information revealed the enormous variability that exists between cancer tumors of different patients: each patient has a cell characterization that is unique only to him and is derived from his characteristic genetic changes. Despite the significant advantage resulting from such precise characterization of the specific patient, this development has led to the fact that comparing different patients is similar to comparing apples and oranges: in the absence of a common denominator, The necessary comparison to find markers associated with prognosis (such as mortality or disease severity) becomes impossible.
The development of the Technion researchers, an algorithm called tuMap, provides an answer to this complex challenge through "difference-based comparison". The innovative algorithm provides a possibility To place many different tumors on a uniform scale, which provides a reference point for comparison. In this way, it is possible to significantly compare tumors of different patients, and even of the same patient over time (eg at diagnosis and after treatment). The algorithm makes it possible to take advantage of the high measurement resolution that is possible today for clinical uses that were not successful before, such as more accurate prediction for various clinical indicators, including tumor recurrence and mortality. Although the researchers tested the algorithm on leukemia tumors, they estimate that it will also be relevant to other types of cancer.
The research was supported by the National Science Foundation, the Rapaport Family Medical Sciences Research Institute and the NIH.
for the article in the journal Cell Systems click here
