The delivery barriers of nonviral vectors have been well documented, yet there are no reports of a rigorous protein by protein analysis of the rate limiting steps. Such an analysis requires a parallel quantification of the effect of each protein on nonviral gene transfer. One approach is to use RNA interference (RNAi) to knock down individual genes, and determine the effect of the corresponding protein. We have developed a RNAi based high throughput screen to probe the effects of 5,520 genes on nonviral gene transfer. The screen consists of reverse transfecting three pooled siRNAs per gene into human endothelial cells at a total concentration of 30 nM followed by the transfection of luciferase plasmid. For quality control, positive controls targeting the luciferase plasmid and scrambled negative controls were included in the screen. The average of the negative controls fell within 1 standard deviation of the median of the sample population, while the average of the positive control was greater than 3 standard deviations from the median. A comparison of transfection reagents revealed that 1:3 (mass ratio plasmid to lipid) siPort NeoFX gave a considerably higher siRNA transfection compared to other transfection reagents, while Lipofectamine 2000 was effective as a model cationic lipid transfection agent for the luciferase plasmid. Those siRNAs that enhance the luminescence signal correspond to potential rate limiting proteins (positive hits), while those siRNAs that reduce signal correspond to proteins potentially critical to the gene transfer mechanism (negative hits). Positive hits and negative hits were selected with a 2.5 standard deviation cutoff above and below the population median, respectively. Approximately 2% of the siRNAs screened were identified as hits, including several siRNAs corresponding to proteins in the solute carrier family found in the endosome. Secondary and tertiary screens with RT-PCR were used to confirm and validate the activity of each hit. In this work, we have applied this RNAi screen to human endothelial cells, which are refractory to nonviral gene transfer due to low mitotic rates in vivo. However, this screen can be applied to other nonviral gene transfer reagents such as novel cationic lipids and polymers to more accurately model the gene transfer mechanism and for the intelligent design of vectors for a given application.