Comparison of genotypic-based microbial source tracking methods requiring a host origin database

Microbial source tracking (MST) results, obtained using identical sample sets and pulsed field gel electrophoresis (PFGE), repetitive element PCR (rep-PCR) and ribotyping techniques were compared. These methods were performed by six investigators in analysis of duplicate, blind sets of water samples spiked with feces from five possible sources (sewage, human, dog, cow and seagull). Investigators were provided with samples of the fecal material used to inoculate the water samples for host origin database construction. All methods correctly identified the dominant source in the majority of the samples. Modifications of some of these methods correctly identified the dominant sources in over 90% of the samples; however, false positive rates were as high as 57%. The high false positive rates appeared to be indirectly proportional to the levels of stringency applied in pattern analysis. All the methods produced useful data but the results highlighted the need to modify and optimize these methods in order to minimize sources of error.