Three steps towards comparability and standardization among molecular methods for characterizing insect communities

Abstract

Molecular methods are currently some of the best-suited technologies for implementation in insect monitoring. However, the field is developing rapidly and lacks agreement on methodology or community standards. To apply DNA-based methods in large-scale monitoring, and to gain insight across commensurate data, we need easy-to-implement standards that improve data comparability. Here, we provide three recommendations for how to improve and harmonize efforts in biodiversity assessment and monitoring via metabarcoding: (i) we should adopt the use of synthetic spike-ins, which will act as positive controls and internal standards; (ii) we should consider using several markers through a multiplex polymerase chain reaction (PCR) approach; and (iii) we should commit to the publication and transparency of all protocol-associated metadata in a standardized fashion. For (i), we provide a ready-to-use recipe for synthetic cytochrome c oxidase spike-ins, which enable between-sample comparisons. For (ii), we propose two gene regions for the implementation of multiplex PCR approaches, thereby achieving a more comprehensive community description. For (iii), we offer guidelines for transparent and unified reporting of field, wet-laboratory and dry-laboratory procedures, as a key to making comparisons between studies. Together, we feel that these three advances will result in joint quality and calibration standards rather than the current laboratory-specific proof of concepts. This article is part of the theme issue 'Towards a toolkit for global insect biodiversity monitoring'.

Keywords: COI synthetic spike-in; comparability; internal standard; metabarcoding; metadata; multiplex.

Figure 1.

Schematic metabarcoding workflow with the proposed attempts of harmonization highlighted in red. While the overall procedure is variable (different substrates, DNA extraction and libraries preparation methods, bioinformatic pipelines), three proposed approaches for improved standardization can be adopted: (i) integration of artificial spike-ins, (ii) consistent marker gene amplification during PCR and consideration of multiplex approaches, and (iii) standardized reporting of metadata. OTU, operational taxonomic unit; ASV, amplicon sequence variant.

Figure 2.

Spike-ins improve quality checks and facilitate normalization. Synthetic spike-ins aid in quality assessment, with deviations in the number and relative proportions of spikes indicating procedural issues (a). When spike-ins yield expected proportions of reads (b) and (c), they can be used for normalization, correcting for amplification and dilutions, and between-sample comparisons.

Figure 3.

The choice of the gene region targeted in metabarcoding studies will impact the insect community detected and can hinder data comparability across studies. (a) Targeting several gene regions using a multiplex approach can be time and cost-effective and (b) targeting the same regions can further allow data comparability between different studies. Animal images in the figure obtained from the Integration and Application Network, University of Maryland Center for Environmental Science (ian.umces.edu/symbols/).

Figure 4.

Complete and well reported metadata (b) strengthens comparability of different metabarcoding studies and enables meta-analysis. Incomplete and non-published documentation of metadata from metabarcoding community analysis (a) and (c) prevents comparability of datasets, hampering meta-analysis and large-scale synthesis as well as the integration in applied monitoring schemes and conservation. Raw data: link to archive (such as NCBI Sequence Read Archive or EBI-ENA), where raw sequence data are stored also including sample description; lab protocol: laboratory protocol information based on the MIMARKS reporting standards (see main text, [72]) and extended in the electronic supplementary material, table S2—laboratory protocol); bioinformatic protocol: minimum reporting standards of main programs and parameter settings for bioinformatic sequence analysis as summarized in the electronic supplementary material, table S2—bioinformatic analyses; geospatial, temporal and environmental metadata: all sample associated metadata as proposed in MIMARKS standards and associated environments (e.g. through GEOME platform, see main text).