• Image Analysis
  • Single Cell / Single Molecule

Artificial intelligence for phenotyping in high-content screens by video-microscopy

Project lead by  Thomas Walter
Industrial partners  CAIRN Biosciences
A

Abstract

The ability of cancer cells to develop resistance to therapy is now a major issue for the development of new treatments in oncology. Cancer cells naturally develop resistance to therapy due to their properties of genetic instability and rapid proliferation. With 17 million new cancer patients diagnosed each year and more than 100 clinically proven targeted oncology therapies, there is a critical need for new drugs to overcome this acquired resistance to cancer treatments.

Our project focuses on PARP inhibitors (PARPi) that target DNA damage repair mechanisms, a key vulnerability for acquired resistance to treatments. Cairn Biosciences generates high throughput live image screening data to profile signaling pathways and phenotypic changes associated with PARPi resistance. The CBIO is a reference laboratory for machine learning applied to biology.

The collaboration proposed in this project will aim at :

1) Develop machine learning (ML) methods for compound classification and phenotypic profiling at the cellular level,
2) use the conceptual framework of causal inference to find causal relationships between vectors of biological characteristics derived from phenotypic profiling analysis,
3) develop ML models to predict cellular biological states from non-labelled in vitro images.

The success of this project will allow the identification of targets and drugs to treat PARPi resistance. In the longer term, our platform puts us in a good position to treat resistance acquired in other therapeutic areas beyond PARPi.

C

Call

As a response to the : Call for projects 2020 : Paris Region PhD²

Paris Region PhD² : call for applications for the funding of PhD grants

Details & Selected Projects
T

Teams

  • Centre for computational biology (CBIO)

    MINES ParisTech
    PSL University

    Read more