Project Details
Description
The wealth of data that can be generated from a clinical sample is rapidly increasing with technological advancements. In the era of immunotherapy, the complex composition of tumor microenvironments needs to be understood in order to make optimal treatment selection for individual patients based on biological insight. Collection of spatial molecular data are thus pivotal, but frameworks for integrative analysis of data from different modalities have yet to be developed.
Goal: We will generate workflows for extracting and integrating tissue image-derived spatial metrics with spatial omics, genomic data, and clinical metadata, to facilitate the development of integrated prediction models to guide choice of therapy.
Specifically, we aim to develop:
1) Spatial metrics - Evaluate deep learning-based workflows for segmentation of tissue images followed by cell classification and extraction of cell-specific distances and structure metrics for describing the tissue microenvironment.
2) Multi-omics data integration – Frameworks for integrating spatial proteomics and transcriptomics collected from multiple regions of interest (ROIs) per tissue biopsy, with image-derived spatial metrics, clinical metadata and patient-matched molecular data (e.g. genomics) retrieved from other studies or as part the diagnostic workup.
3) Spatial and multi-omics prediction models – Machine learning-based prediction models based on condensed sets of features selected from multimodal data, for patient stratification and informed clinical decision-making.
4) Spatial Omics user applications – Integrative, web-based applications embedding image analysis for extracting spatial metrics, spatial and multi-omics data integration and analysis, for interactive processing, visualization, interpretation and sharing of multimodal data.
Goal: We will generate workflows for extracting and integrating tissue image-derived spatial metrics with spatial omics, genomic data, and clinical metadata, to facilitate the development of integrated prediction models to guide choice of therapy.
Specifically, we aim to develop:
1) Spatial metrics - Evaluate deep learning-based workflows for segmentation of tissue images followed by cell classification and extraction of cell-specific distances and structure metrics for describing the tissue microenvironment.
2) Multi-omics data integration – Frameworks for integrating spatial proteomics and transcriptomics collected from multiple regions of interest (ROIs) per tissue biopsy, with image-derived spatial metrics, clinical metadata and patient-matched molecular data (e.g. genomics) retrieved from other studies or as part the diagnostic workup.
3) Spatial and multi-omics prediction models – Machine learning-based prediction models based on condensed sets of features selected from multimodal data, for patient stratification and informed clinical decision-making.
4) Spatial Omics user applications – Integrative, web-based applications embedding image analysis for extracting spatial metrics, spatial and multi-omics data integration and analysis, for interactive processing, visualization, interpretation and sharing of multimodal data.
| Short title | eSSENCE@LU 9:4 |
|---|---|
| Status | Finished |
| Effective start/end date | 2023/01/01 → 2024/12/31 |
UN Sustainable Development Goals
In 2015, UN member states agreed to 17 global Sustainable Development Goals (SDGs) to end poverty, protect the planet and ensure prosperity for all. This project contributes towards the following SDG(s):
