The aggregation of tau protein into paired helical filaments is one of the hallmarks of Alzheimer's disease and related dementias. We therefore continue our search for non-toxic, cell penetrating inhibitors of tau aggregation, which hold potential for brain penetration. Pickhardt et al. (2005) have reported a high throughput screen for tau aggregation inhibitors previously, which resulted in the identification of several hit classes. Here we report the identification of novel inhibitors which were not present in the initial high throughput assay. This was achieved by transformation of the high throughput screen data into the 3D relationships of virtual pharmacophores The pharmacophore models were utilized in a virtual screen of a Maybridge database. The virtual screen provided 136 hits; 19 representative hits were selected and assayed, this resulted in two novel leads with an IC(50) < 13 microM. These two leads feature a novel scaffold for tau aggregation inhibitors.