Homogenization of insect bulk samples yields more comprehensive yet comparable biodiversity data than non-destructive lysis

The rapid advancement of molecular biodiversity monitoring tools, particularly DNA metabarcoding, has improved specimen identification in bulk samples, such as those from Malaise traps, where traditional morphological identification is impractical. While not yet standardized, a typical first step in insect bulk sample analysis is the extraction of DNA from homogenized specimens. While this step yields reliable metabarcoding results, it destroys the specimens, preventing further use in monitoring and taxonomic analysis. Non-destructive lysis, which preserves specimen integrity, is still being evaluated for its effectiveness in accurately assessing bulk sample biodiversity. In this study, we assessed the suitability of non-destructive lysis for Malaise trap samples and compared its performance with homogenization using an established metabarcoding workflow. Five bulk samples were collected with Malaise traps. Samples were first incubated in a lysis buffer containing Proteinase K (non-destructive lysis) and then homogenized. DNA was extracted from both treatments and metabarcoding was performed to compare OTU richness, accumulation, and beta diversity. On average, homogenized samples yielded 3.8% more OTUs than non-destructive lysis samples. Although homogenization provides a more comprehensive and cost-effective assessment of Malaise trap bulk samples, non-destructive lysis still recovered at least 80% of the OTUs identified through homogenization and revealed similar patterns of community change. Even though our results show that both methods yield comparable data on insect biodiversity and can be used for monitoring, we consider non-destructive lysis as not suitable for integration into automated workflows or large-scale biomonitoring due to the much higher costs. Nonetheless, this method remains important in cases where morphological integrity needs to be preserved and additional sampling is not possible.