1005 Alcyon Dr Bellmawr NJ 08031
+1 484-446-2566
support@bmcpros.com
To produce effective targeted therapies for cancer, scientists need to isolate the genetic and phenotypic characteristics of cancer cells within and across different tumors, because these differences can affect the tumor’s response to treatment.
Part of this work requires a deeper understanding of the RNA or protein molecule of each cancer cell located in the tumor and how it looks under a microscope.
Traditionally, scientists have studied one or more of these aspects separately, but now, using a new deep learning AI tool, CellLens (Cell Local Environment and Neighborhood Scan), to fuse all three fields together and combine a combination of convolutional neural networks and graphical neural networks to build a comprehensive digital profile for each unit. This allows the system to group cells with similar biology – even cells that look very similar in isolation, behave differently according to the surrounding environment.
This study was recently published in Natural Immunologya detailed introduction to the results of a collaboration between researchers at MIT, Harvard Medical School, Yale, Stanford and the University of Pennsylvania, led by Bokai Zhu, a postdoctoral fellow at MIT, MIT, MIT and Harvard Extensive Institute, and the MGH Ragon Institute of MGH, MIT, MIT and MIT and Harvard University.
Zhu explained the impact of this new tool: “Initially we would say, oh, I found a cell. This is called a T cell. Using the same dataset, using cells by applying them, now I can say that this is a T cell that is currently attacking a specific tumor boundary in patients.
“I can use the existing information to better define what a cell is, what the subpopulation of that cell is, what the cell is doing, and what the cell is potentially functional readings. The method can be used to identify new biomarkers that provide specific and detailed information about the diseased cells that can be done more targeted therapeutic development.”
This is a key advance because current methodology often misses key molecular or contextual information, for example, immunotherapy may target cells that are only present on the tumor boundaries, thus limiting efficacy. By using deep learning, researchers can use cell leans to detect many different layers of information, including morphology and the location of cells in the tissue.
When applied to samples of healthy tissues and several types of cancers, including lymphoma and liver cancer, Celllens discovered rare subtypes of immune cells and revealed the relationship between their activity and location and disease processes—such as tumor infiltration or immunosuppression.
These findings could help scientists better understand how the immune system interacts with tumors and pave the way for more precise cancer diagnosis and immunotherapy.
“I’m extremely excited by the potential of new AI tools, like CellLENS, to help us more holistically understand aberrant cellular behaviors within tissues,” says co-author Alex K. Shalek, the director of the Institute for Medical Engineering and Science (IMES), the JW Kieckhefer Professor in IMES and Chemistry, and an extramural member of the Koch Institute for Integrative Cancer Research at MIT, as well as an Institute member of the Great Institute and the Lagon Institute. “We can now measure a lot of information about individual cells and their tissue environments through cutting-edge, multi-friction assays. Effectively leveraging this data to nominate new therapeutic potential customers is a critical step in developing improved interventions. When combined with the correct input data and carefully verified, such tools can relate to our ability to make our health and wellness and wellness and wellness and wellness.