One of the principal reasons for the high costs of drug development is the failure in the toxicity and/or efficacy stages when a drug is tested in vivo. Fly models can be used for screening entire libraries of compounds, for validating lead therapeutics quickly and effectively, and for testing drug toxicity and efficacy in vivo for various human diseases. We have developed a screen for neuroprotective compounds in our ALS models (Patent Application No. 14/508,610, pending) and screened both 1,200 FDA approved compounds. Among these we found that pioglitazone, a drug approved by the FDA to treat diabetes mellitus can improve locomotor function but does not extend lifespan in ALS flies. A similar result was reported in clinical trials, which showed that pioglitazone has no significant improvement in survival despite promising results in a SOD1 mouse model of ALS. In retrospect, the results in the fly model were similar to the results of the human trials, suggesting that perhaps fly models could provide a better, more cost-effective and rapid prediction of outcome in clinical trials. In addition, we are testing “new chemical matter” compounds synthesized by our collaborator, Dr. Jon Njardarson (Chemistry and Biochemistry, University of Arizona) and identified four small molecules with neuroprotective potential in vivo. We are currently validating these small molecules in the fly model and plan to test them in a mammalian model in the near future. We are also collaborating with biotech companies to screen compounds of interest.