We introduce magnetophoresis-based microfluidics for sorting biological targets using positive Magnetophoresis (pM) for magnetically labeled particles and negative Magnetophoresis (nM) for label-free particles. A single, externally magnetized ferromagnetic wire induces repulsive forces and is positioned across the focused sample flow near the main channel's closed end. We analyze magnetic attributes and separation performance under two transverse dual-mode magnetic configurations, examining magnetic fields, hydrodynamics, and forces on microparticles of varying sizes and properties. In pM, the dual-magnet arrangement (DMA) for sorting three distinct particles shows higher magnetic gradient generation and throughput than the single-magnet arrangement (SMA). In nM, the numerical results for SMA sorting of red blood cells (RBCs), white blood cells (WBCs), and prostate cancer cells (PC3-9) demonstrate superior magnetic properties and throughput compared to DMA. Magnetized wire linear movement is a key design parameter, allowing device customization. An automated device for handling more targets can be created by manipulating magnetophoretic repulsion forces. The transverse wire and magnet arrangement accommodate increased channel depth without sacrificing efficiency, yielding higher throughput than other devices. Experimental validation using soft lithography and 3D printing confirms successful sorting and separation, aligning well with numerical results. This demonstrates the successful sorting and separating of injected particles within a hydrodynamically focused sample in all systems. Both numerical and experimental findings indicate a separation accuracy of 100% across various Reynolds numbers. The primary channel dimensions measure 100 µm in height and 200 µm in width. N52 permanent magnets were employed in both numerical simulations and experiments. For numerical simulations, a remanent flux density of 1.48 T was utilized. In the experimental setup, magnets measuring 0.5 × 0.5 × 0.125 inches and 0.5 × 0.5 × 1 inch were employed. The experimental data confirm the device's capability to achieve 100% separation accuracy at a Reynolds number of 3. However, this study did not explore the potential impact of increased flow rates on separation accuracy.