Welcome to DMG-ADAPTS

Our platform provides dedicated pipelines to analyse time-series multiomics data and
online tools to explore tumour adaptations to therapeutics in space & time.

Ready to dive into our tools and datasets?

Explore Now

About us

The DMG-ADAPTS (ADvanced mAchine-learning Precision Treatment Strategy) platform presents a unique temporal multiomics characterisation of on-treatment diffuse midline glioma (DMG) patients and experimental models. This includes single-cell RNA sequencing, single-cell T Cell Repertoire (TCR) sequencing, spatial transcriptomics (Xenium), histological slide images, CUT&RUN assays, proteins, cytokines and chemokines profiling.

Our objective is to model the temporal adaptation of tumours to therapeutics by gathering multiomics data from patients and experimental models undergoing treatment across various time points.



Why it matters

Diffuse midline glioma (DMG), including diffuse intrinsic pontine glioma (DIPG), represents a uniformly fatal group of brain tumours predominantly affecting children, adolescents, and young adults with palliative radiation therapy standing as the solitary treatment option. Over the past decade, therapeutic vulnerabilities in DMG have been targeted using precision strategies but anti-DMG responses have been only temporary. The rapid resistance to treatments in DMGs is driven by molecular and cellular plasticity and mechanisms of immune escape that need to be understood and predicted to enable long-term treatment benefits for patients.

Our Image


Provide a proactive approach to patient care by monitoring tumour adaptation through non-invasive profiling.



Our Image

Our Vision

The DMG-ADAPTS platform is designed to model the dynamic of tumour adaptation to promising therapeutics and predict the best sequential temporal treatment schema based on time-series multiomics profiling data. All collected data serve as the foundation for training DMG-ADAPTS machine learning models, featured to:

  • Recommend the most pertinent treatment strategy for a patient, based on their molecular profile, to target the identified signalling dependency;
  • Adjust therapeutic protocols in real-time, guided by tumour adaptation’s biomarkers (e.g. clonotypes, signalling pathways) through non-invasive liquid biopsies.

These biomarkers will act as surrogates for tumour changes and be monitored all along patient’s journey to indicate the optimal timeframe for repositioning the next therapy.




Explore our Tools

Visualisation

Spatiotemporal exploration of gene expression levels across cell populations and conditions.

DE-Analysis

Interactive and customisable analysis of differential expression between conditions.

Trajectory inference

Advanced tools for deciphering Gene Regulatory Networks and spatiotemporal trajectories.

Copy Number Variation

Study intratumoural heterogeneity by infering Copy Number Variations (CNVs).

Cell-Cell communication

Comprehensive analysis of cellular interactions between clusters with CellPhoneDB.

Pathway
activation

Detect dysregulated pathways through the combined use of GSEA and PROGENY.





Contributing Institutions

Our Image