02 — Map
Turn multi-omic EV observations into tissue-anchored, testable biological programs.
Which molecular programs track progression—and what CNS biology could explain them?
Interpretable multi-omics, biological calibration, and evidence-backed hypothesis generation.
From features to programs
The objective is not a black-box score. Models are organized around coherent miRNA, mRNA, protein, imaging, and clinical relationships so that a candidate signature can be traced back to pathways, cell contexts, and experimental questions.
Loss of coordination across regulatory and effector molecular programs
A working hypothesis is that disease progression may involve a loss of coordination across regulatory and effector molecular programs. Mapping asks where that coordination changes, whether it is robust across cohorts, and whether it is biologically plausible.
A model must explain its route back to biology
The analytical workflow uses program-level relationships rather than isolated rankings: quality-controlled molecular inputs, cohort-aware model development, biologically interpretable features, tissue and cell-context anchoring, and a locked evaluation plan for the next cohort.