Down syndrome cell adhesion molecule 2 (Dscam2) promotes brain wiring primarily through homophilic repulsion; a mechanism with the unusual dichotomy of adhesion followed by repulsion. The signalling molecules that induce homophilic repulsion are currently unknown. We therefore knocked a GFP tag into the endogenous Dscam2 locus and identified binding partners via immunoprecipitation and mass spectrometry. To determine which of the candidates were involved in Dscam2 repulsion, we developed an in vivo repulsion assay. Dscam2 was overexpressed in sensory neurons that normally form tight fascicles in the central nervous system. Repulsion was quantified as the amount of defasciculation observed in their axons. A screen was performed in this background using RNAi to knock down ~50 candidate Dscam2 binding proteins. Knock down of some candidates suppressed and others potentiated the defasciculation phenotype, suggesting that we had identified genes that promote and inhibit Dscam2 repulsion. A large proportion of validated genes were linked to PI3K signalling and our data were consistent with PI3K promoting Dscam2 repulsion. However, contradictory phenotypes were also observed; for example, removing a well characterised promoter of PI3K potentiated Dscam2 repulsion. We hypothesised that knocking down critical regulators of PI3K throughout development could lead to compensatory changes in gene expression. To test this, we used an inducible system to restrict expression to a late developmental window and showed that this corrected the contradictory results. We therefore conclude that PI3K signalling works downstream of Dscam2 homophilic binding to drive repulsion between neurons.