Glioblastoma research

Join our AI moonshot to accelerate open science

Join our research projects to benefit the glioblastoma research and patient community.

Contribute to our key scientific goals

Goal 1

Develop a better understanding of long survivors
  • To identify those most likely to benefit from treatment / most at risk
  • To discover new biology of long survivors

AI capabilities & datasets

histology-based survival predictive models

Develop histology-based survival predictive models, using Owkin’s multiple-instance learning algorithm1

AI ready

Use interpretable AI models to identify associated histological biomarker(s)


Large dataset (>500 patients) from APHP2 that includes long survivors, with clinical, H&E, and molecular data

Goal 2

Identify novel therapeutic targets in glioblastoma
  • Target identification using novel data modalities (e.g. spatial omics)
  • Associate the targets to novel, spatially-defined glioblastoma subtypes

AI capabilities & datasets

Targets and subgroups

Co-optimization and prioritization of targets and subgroups using end-to-end machine learning methods

team of biologists

In-house team of biologists and pharmacologists, and wet lab capabilities to validate ML findings


Multimodal and spatial dataset (100+ patients, 6 modalities) from Owkin’s MOSAIC project

(1) Pierre Courtiol, et al. “Deep learning-based classification of mesothelioma improves prediction of patient outcome” Nat Med 25, 1519–1525 (2019)

[2] Assistance Publique–Hôpitaux de Paris (AP-HP) is the university hospital trust operating in Paris and its surroundings. With 38 hospitals, it is the largest hospital system in Europe and one of the largest in the world.

Mission milestones

May to June 2024
Get data AI-ready
June to November 2024
AI model development and research
November 2024 to April 2025
Target validation (wet lab)
Groundbreaking publication
We aim to share the results of this research with the scientific community through submission for publication in peer-reviewed scientific journals.