The main aim of ESR14 project will be to address the CKD heterogeneity through data-driven stratification approaches to open up new opportunities for personalized drug discovery. The CKD patient population includes a wide range of diverse aetiologies with a multitude of underlying molecular processes in the kidney. Moreover, the kidney is structurally heterogeneous and comprised of more than 25 different cell types impeding interpretation of whole-tissue omics profiling.
The specific objectives of ESR14 will be:
- using available kidney transcriptomics data, identify inherent patient molecular subtypes in the CKD population using unsupervised machine learning clustering approaches (k-means, DBSCAN, neural network);
- perform computational deconvolution of cell-specific signals within the renal tissue transcriptomics data using CellMix algorithms, reconcile with the population stratification findings and perform pathway enrichment analyses to reveal the underlying molecular processes;
- per patient subtype, identify candidate drug targets and biomarkers through unbiased data-driven (using statistical feature selection) and knowledge-based (using pre-existing biological and druggability insights) gene prioritization approaches. Ultimately, the scope of ESR14 project will be to validate the targets preclinically using in vivo suitable tool compounds in translatable animal models in collaboration with the other ESRs.