Bioinformatics for Biomonitoring: Species Detection and Diversity Estimates Across Next-Generation Sequencing Platforms

摘要: Abstract As a fast-growing area of technology, sequencing platforms are updated frequently and this rapid technical revolution poses not only great advances but also challenges. To be effective, biomonitoring programmes need to deliver comparable results across research groups time. Understanding the sources bias in bioinformatics promotes reliable that accurately reflect biodiversity. We assembled two mock communities planktonic organisms assess accuracy species recovery based on 18S rRNA V4 region using NGS platforms, Roche 454 (the platform choice for early metabarcoding studies), Illumina MiSeq (employed recent studies). Our findings suggest have performance datasets. When singletons (sequences represented by single read) were excluded from analyses, had slightly better operational taxonomic unit (OTU) precision score than (calculated as number detected divided OTUs generated) one workflow (when paired reads appended, merged). performed terms detection when simple with individual per analysed. singleton sequences included, both more 75% higher achieved MiSeq. The OTU clustering datasets resulted gross overestimation richness. This finding suggests studies employing proxy genetic diversity must carefully perform read processing, such exclusion, avoid overestimates. Finally, study provides insight into bioinformatic strategies should accompany transitions. In field metabarcoding, where technology constantly drive discipline, ensuring comparability past future technologies, derived ecological conclusions is important.