Replacing Sanger with Next Generation Sequencing to improve coverage and quality of reference DNA barcodes for plants

Abstract

We estimate the global BOLD Systems database holds core DNA barcodes (rbcL + matK) for about 15% of land plant species and that comprehensive species coverage is still many decades away. Interim performance of the resource is compromised by variable sequence overlap and modest information content within each barcode. Our model predicts that the proportion of species-unique barcodes reduces as the database grows and that ‘false’ species-unique barcodes remain >5% until the database is almost complete. We conclude the current rbcL + matK barcode is unfit for purpose. Genome skimming and supplementary barcodes could improve diagnostic power but would slow new barcode acquisition. We therefore present two novel Next Generation Sequencing protocols (with freeware) capable of accurate, massively parallel de novo assembly of high quality DNA barcodes of >1400 bp. We explore how these capabilities could enhance species diagnosis in the coming decades.

Incomplete estimates of genetic diversity within species: Implications for DNA barcoding

“…Wilkinson et al. () points to strong ascertainment bias as the most likely explanation for this. In the early days of BOLD, DNA barcode sequence acquisition was high, due to the fact that over 75% of taxon records were mined from already well‐established sequence databases such as GenBank (Wilkinson et al., ).…”

“…Sampling (ascertainment) bias is an important factor to consider in this regard because it can lead to under‐ or overestimation of population parameters. Ascertainment bias describes the tendency of certain individuals to be less likely sampled than others (Parr, Guralnick, Cellinese, & Page, ) and is common in molecular biodiversity studies (Hanner, Becker, Ivanova, & Steinke, ; Muirhead et al., ; Mutanen et al., ; Wilkinson et al., ). This can occur, for example, when sampling is restricted to certain geographic regions (Muirhead et al., ) or to particular species (e.g., those known to be of conservation importance) (Hanner et al., ).…”

Dropping Hints: Estimating the diets of livestock in rangelands using DNA metabarcoding of faeces

“…In that case, when one searches the barcode library with a query sequence, the closest matching sequence in the library (even if it is an exact match for a slowly evolving gene like rbcL) may not be the same species, or even a congeneric species. Indeed, Wilkinson et al (2017) caution that the proportion of species-unique rbcL + matK barcodes is likely to drop even further as sampling increases. To make a judgement call about species identity, one has to know what closely related species are found in the sampling area, and whether those species are represented in your reference library.…”

“…First introduced by Straub et al (2011) in the context of plant systematics, genome skimming provides much more sequence data than barcoding, and might represent a new direction in identifying plant samples from molecular data , Wilkinson et al 2017). …”

Forensic DNA Technological Advancements as an Emerging Perspective on Medico-Legal Autopsy: A Mini Review

“…chemotyping [233,234], and even the estimation of postmortem interval and skin microbiomes [235,236]. Other various aspects of possible nonhuman biological evidences found on the corpse, e.g., pollens or plant fragments, can reveal potential indirect links to the perpetrator, and also additional, case-related facts, complementing autopsy information with the modality of death or the crime scene setting for purposes of case reconstruction [237,238]. The human-related techniques of forensic genetics have been adapted to nonhuman analyses, but the genetic markers implemented frequently originate from the field of conservation biology.…”