An exploration of DNA extraction methods of fly iDNA for a scalable biodiversity monitoring

Background Metabarcoding of invertebrate-derived DNA (iDNA) has proven to be an excellent tool for assessing terrestrial mammal diversity. While fly iDNA-based monitoring can detect a wide range of mammals, including rare, small, and cryptic species, a major constraint to wider adoption has been the time and costs associated with sample processing. Aims and Methods Here we explored the effectiveness of bulk fly iDNA extraction methods which increase the pooling of flies by an order of magnitude over currently used extraction methods. One extraction method involved removing a single leg per fly from many different flies, pooling these legs, and destructively extracting DNA from the pool (Nflies legs per pool = 105). The other method involved non-destructively extracting DNA from a large pool of entire flies (Nflies per pool = 105). We tested these methods on flies collected at eight sites, representing three different habitats across five countries in sub-Saharan Africa (Nflies/site = 105; Nflies total = 840). We compared the mammal species detected using metabarcoding of these extracts with extracts generated using an approach that is currently widely used in the field, the cost- and labor-intensive destructive extraction of small pools of entire flies (Nflies per pool = 7; Nflies per site = 105 flies). Results The non-destructive extraction from large pools of entire flies detected a greater number of mammals (Nspecies detections total = 59; x̄species/site = 7.4; range = 1 to 12) than the destructive extraction from large pools of fly legs (Nspecies detections total = 15; x̄species/site = 1.9; range = 2 to 3), while detecting a similar number of mammals as the destructive extraction from small pools of entire flies (Nspecies detections total = 67; x̄species/site = 8.4; range = 5 to 15). Conclusion Our findings indicate that the non-destructive extraction of large pools of entire fly bodies has the potential to streamline and reduce costs of fly iDNA extraction. We hope this will further promote the use of fly iDNA in large-scale terrestrial biomonitoring efforts.