Launch Of Deepcell’s AI-Powered Single Cell Analysis Platform To Speed Up Cell Biology Research And Advance Morpholomics

The REM-I Platform, a high-dimensional cell morphology analysis and sorting platform that consists of the REM-I benchtop instrument, Human Foundation Model, and Axon data suite, has been launched by Deepcell, a pioneer in AI-powered single cell analysis to fuel deep biological discoveries. The REM-I Platform will stimulate new ways of discovery in a variety of domains, including cancer biology, developmental biology, stem cell biology, gene therapy, and functional screening, among others, by combining single cell imaging, sorting, and high-dimensional analysis. During three scientific podium presentations at CYTO 2023 in Montreal, Quebec, May 20–24, 2023, the executives of Deepcell will share data on the company’s AI-based morphology profiling solutions, which make use of the company’s unique deep learning and computer vision models.

“Deepcell’s approach to bringing artificial intelligence into cellular analysis will revolutionise biological research, ushering in a new era of discovery,” said Maddison Masaeli, PhD, cofounder and CEO of Deepcell. “By applying the most recent developments in AI to morphology, which is the cornerstone of cell biology,” the company says, “we empower our customers to quickly transform biological research.”

Since the invention of the microscope, one of the first ways cells were investigated was through their morphology. Despite recent improvements in flow cytometry and microscopy, the area of cell biology hypothesis has remained constrained and dependent on interpretation up until now due to the lack of adequate instruments for cellular measurement and characterization. Cell morphology can finally join other high-dimensional, single-cell analysis methodologies with the latest generation of AI and machine learning models, including Deepcell’s Human Foundation Model, allowing researchers to fully utilise the morpholome.

Euan Ashley, MD, PhD, scientific cofounder of Deepcell, associate dean in the Stanford University School of Medicine, and professor at Stanford University, said: “In the launch of the REM-I Platform, we are witnessing the realisation of years of first-principle thinking about the future of cell biology—a future liberated from the constraints of prior knowledge.” “With the aid of powerful artificial intelligence models, we can look progressively deeper into the biology of individual cells and go beyond the limitations of what human eyes can see. I eagerly anticipate what the scientific community will do with this potent new weapon.

With the use of Deepcell technology, more than two billion images of single cells from a wide range of cell types have been captured and characterised. Brightfield single cell images acquired with the REM-I instrument are characterised by The Human Foundation Model, a self-supervised deep learning model trained on a subset of these unlabeled cellular images from a variety of carefully chosen biological samples, which also produces high-dimensional embedding data. Researchers can sort their target cell groups into up to six outlets on the REM-I instrument using the Axon data suite to retrieve, visualise, and analyse this data in real time.

“Up until this point, the study of morphology was restricted to the interpretation of cellular characteristics by humans. To scale up and democratise the collection of single cell data and to enable unprecedented insights, morphology-powered discovery must be advanced, according to Mahyar Salek, PhD, cofounder, president, and chief technology officer at Deepcell. Similar to how next-generation sequencing changed how we understood the genome, advances in machine learning will revolutionise how we understand cell phenotypes.

The Translational Genomics Research Institute and the University of California, San Francisco, joined up with Deepcell to develop the Technology Access Programme, which used the technology to analyse human cell lines, body fluids, and solid tissues as part of cancer research and drug screening programmes.

Through its Technology Access Programme, the business just finished installing Deepcell technology for the first time in Europe at the Erasmus Medical Centre in Rotterdam, where it will be used to research immune treatments using cancer patient samples.

According to Peter van der Spek, professor in the Department of Pathology and Clinical Bioinformatics at Erasmus Medical Centre, “The Deepcell platform gives us the ability to discriminate between activated and naive T cells and provides next-level detection of therapy response in peripheral blood mononuclear cells derived from patients treated with immune therapies for cancer.” “Pathologists can examine many more cells than by traditional light microscopy and increase the throughput of samples.”