The successes of cancer research stem from the identification of oncogenes and tumour suppressor genes and the study of their pathways in animal and cell line models. Confronting these models with ‘real life’ human cancer has often been a challenge. This grant aims at integral understanding of the aggressive childhood tumor neuroblastoma. Starting point is a huge database for this tumor. We established mRNA, miRNA, CGH and SNP profiles for a series of neuroblastoma, and sequenced their whole genomes. Over 1500 mRNA profiles of cell lines with manipulated gene expression complete this resource. A novel bioinformatic platform integrated all molecular and clinical data. Several research topics have emerged: (1) Recurrent mutations were detected in neuronal growth cone genes. We will investigate whether failed neuritogenesis is oncogenic; (2) Contrary to the paradigm that cancer is caused by gene mutations, our data suggest that human tumours equally much result from chromosomal gains and losses. They proportionally change the expression of hundreds of otherwise intact genes. We will investigate how copy number changes activate oncogenic pathways. (3) Each neuroblastoma includes two cell types, which at low frequency convert into each other. They differ in mesenchymal and neuro-epithelial character, cancer pathway activation, motility and drug sensitivity and thereby are at the crossroads of cancer stem cell-, drug resistance- and metastasis-research. Transgene expression could induce transitions between both cell types in vitro, which will allow us to elucidate the wiring of both states. (4) Neuroblastoma can go in remission upon treatment, but often relapse as fatal therapy resistant tumour. This plasticity may stem from resistant mesenchymal cells, inter-conversion of cell types and continued chromosomal copy number changes. We will target pathways essential to each cell type to contain plasticity and identify therapeutic options to prepare for clinical testing.