High-grade gliomas are associated with very poor survival. Aggressive brain tumor cells invade into the surrounding normal brain and are impossible to remove by surgery. These cells are protected inside the brain from cancer drugs, leading inevitably to tumor regrowth and ultimately, patient death. There is an urgent need to develop therapeutics that can effectively enter the brain and target cancer cells. A protein called B/bΔg is an ideal target because it is present only in glioma cells, and is absent from normal brain cells. One-bead-one-compound (OBOC) libraries, which are composed of tens of thousands to millions of beads, each coated with a unique compound are widely used tools for discovering novel targeting ligands. In a search for glioma-specific peptides, we have screened an OBOC peptide library by labeling positive ‘hit’ beads with small magnetic particles coated with a B/bΔg-derived peptide. To isolate the magnetized ‘hit’ beads, we have developed a cost-effective and reproducible microfluidic magnetic-activated bead sorter. We have demonstrated that this device can rapidly sort magnetized OBOC beads with high throughput (15,000 beads per hour), specificity (>96%) and sensitivity (>99%). We have isolated several hundred ‘hit’ beads that were then subjected to a secondary screen using B/bΔg-overexpressing cells. Using these strategies, we have isolated and identified 8 novel B/bΔg-targeted peptides. All 8 peptides displayed increased uptake by B/bΔg-overexpressing cells compared to control cells. This increase in cellular uptake was not observed when control peptides were used. Circular dichroism analyses and competition studies revealed that 3 of these peptides displayed specific binding to the B/bΔg-derived peptide. Binding kinetics analyses and in vivo tests are currently underway. These peptides have great potential to be developed into next-generation therapeutic agents against high-grade gliomas that can be translated to improve outcome and quality of life in brain cancer patients.