"It was my pleasure working with Alexander Rakhlin and Sergey Nikolenko, true deep learning experts who made it work."

Theodore Alexandrov

Research Team Leader

Client: EMBL

Founded in 1974, EMBL is Europe’s flagship laboratory for the life sciences – an intergovernmental organisation with more than 80 independent research groups covering the spectrum of molecular biology. It operates across six sites: Heidelberg, Barcelona, Hamburg, Grenoble, Rome and EMBL-EBI Hinxton.

Industry

Healthcare

The SCIENCE Challenge

Researchers at EMBL sought to enhance traditional microbiology methods with Deep Learning in order to reconstruct the complex biological phenomenon that underpin the life cycle of cells.

embl image

Solution

Custom Deep Learning in Computer Vision
Developed and deployed on Neu.ro Platform

Results

DL insights into complex biological processes
Successful reconstruction of the cell lifecycle clock
Potential applications in cancer research

The SOLUTION

An important new healthcare application has been developed on the Neu.ro Platform in coordination with the EMBL. The EMBL led collaboration with scientists around the world to create DeepCycle – an AI-driven system with potential applications in cancer research.

EMBL worked with global AI researchers including Neuromation Chief Research Officer, Sergey Nikolenko, and Senior AI Researcher, Alexander Rakhlin, to develop DeepCycle, an AI-driven technology that models the lifecycle of cells – how they grow and divide.

Using approximately 2.6 million microscopy images of canine kidney cells, the novel deep learning model is able to reconstruct complex biological phenomena based solely on visual data.

Sergey Nikolenko says: “This has been a large and very interesting project on a state of the art topic in bio- informatics: analysis of the cell cycle based on microscopy images. It has been a multidisciplinary effort, but from the AI/ML side, for the first time ever, we have been able to develop distributed representations of cell images that actually have a closed cell cycle progression in time. These representations can be used to identify the “cell clock”, i.e., current “age” of a cell, which may have important implications across the medical field.”

“It’s a pretty advanced use of deep learning. Rather than using any existing model, we came up with a strategy to make a deep learning model to learn about subtle differences of the cells. This ultimately led to the model being able to reconstruct the cell cycle trajectory. Intriguingly, the model learned such a complex phenomenon by itself – just from how cells look.”

– Theodore Alexandrov