How this graph builds itself

Why chromosomes break and rejoin at all, whether fusion is ever intended biology, and whether we could intervene before the fusion exists.

Exon-level junctions for every partner, which domains survive each break, and whether different breaks mean different disease.

The causal chain from a hybrid protein to a blood cell that will not mature.

Fetal liver versus marrow, the primed progenitor, and lineage switching under treatment.

Every contact surface the fusion depends on, with structures, affinities, essentiality and toxicity.

What has been tried, what is in trials, and every published way the disease escapes.

Presentation, diagnosis, treatment course and real published patient accounts.

How many children, what care costs, and why one mechanism can serve many rare populations.

Who holds which piece of the problem and who is worth partnering with.

Co-mutations, enhancer hijacking, long non-coding RNAs and what else tends to go wrong alongside the fusion.

Every pediatric leukemia fusion family side by side, plus solid-tumour comparators.

Patterns borrowed from finance, law, engineering and ecology, each required to produce a testable prediction.

Reproducible scripts against public data, each producing an entry in the graph.

For each hypothesis: the druggability dossier, datasets, models and go/no-go criteria.