Limits...
DNA barcoding the native flowering plants and conifers of Wales.

de Vere N, Rich TC, Ford CR, Trinder SA, Long C, Moore CW, Satterthwaite D, Davies H, Allainguillaume J, Ronca S, Tatarinova T, Garbett H, Walker K, Wilkinson MJ - PLoS ONE (2012)

Bottom Line: We have sampled multiple individuals for each species, resulting in 3304 rbcL and 2419 matK sequences.These approaches yield similar results, providing relative discrimination levels of 69.4 to 74.9% of all species and 98.6 to 99.8% of genera using both markers.Species discrimination can be further improved using spatially explicit sampling.

View Article: PubMed Central - PubMed

Affiliation: National Botanic Garden of Wales, Llanarthne, United Kingdom. natasha.devere@gardenofwales.org.uk

ABSTRACT
We present the first national DNA barcode resource that covers the native flowering plants and conifers for the nation of Wales (1143 species). Using the plant DNA barcode markers rbcL and matK, we have assembled 97.7% coverage for rbcL, 90.2% for matK, and a dual-locus barcode for 89.7% of the native Welsh flora. We have sampled multiple individuals for each species, resulting in 3304 rbcL and 2419 matK sequences. The majority of our samples (85%) are from DNA extracted from herbarium specimens. Recoverability of DNA barcodes is lower using herbarium specimens, compared to freshly collected material, mostly due to lower amplification success, but this is balanced by the increased efficiency of sampling species that have already been collected, identified, and verified by taxonomic experts. The effectiveness of the DNA barcodes for identification (level of discrimination) is assessed using four approaches: the presence of a barcode gap (using pairwise and multiple alignments), formation of monophyletic groups using Neighbour-Joining trees, and sequence similarity in BLASTn searches. These approaches yield similar results, providing relative discrimination levels of 69.4 to 74.9% of all species and 98.6 to 99.8% of genera using both markers. Species discrimination can be further improved using spatially explicit sampling. Mean species discrimination using barcode gap analysis (with a multiple alignment) is 81.6% within 10×10 km squares and 93.3% for 2×2 km squares. Our database of DNA barcodes for Welsh native flowering plants and conifers represents the most complete coverage of any national flora, and offers a valuable platform for a wide range of applications that require accurate species identification.

Show MeSH
Ability of the DNA barcode markers rbcL and matK to discriminate the Welsh flora.Discrimination (%) at species, genus and family level for rbcL, matK and both markers combined using monophyletic groups in Neighbour-Joining trees (Tree), BLASTn searches (BLAST) and barcode gap analysis using pairwise (Barcode gap pairwise) and multiple alignments (Barcode gap multiple). Species level discrimination for monophyletic groups in Neighbour-Joining trees is divided into bootstrap support values of ‘any’, >50% and >70% based on 1000 bootstrap replicates. Discrimination uses 808 species for which multiple individuals were DNA barcoded for both rbcL and matK. Species with single sequences were included in the analyses as sources of discrimination failure. For a complete list of which species can be discriminated using the different methods see Dataset S3.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3368937&req=5

pone-0037945-g003: Ability of the DNA barcode markers rbcL and matK to discriminate the Welsh flora.Discrimination (%) at species, genus and family level for rbcL, matK and both markers combined using monophyletic groups in Neighbour-Joining trees (Tree), BLASTn searches (BLAST) and barcode gap analysis using pairwise (Barcode gap pairwise) and multiple alignments (Barcode gap multiple). Species level discrimination for monophyletic groups in Neighbour-Joining trees is divided into bootstrap support values of ‘any’, >50% and >70% based on 1000 bootstrap replicates. Discrimination uses 808 species for which multiple individuals were DNA barcoded for both rbcL and matK. Species with single sequences were included in the analyses as sources of discrimination failure. For a complete list of which species can be discriminated using the different methods see Dataset S3.

Mentions: In order to compare across markers and methods of discrimination, we used a dataset of 808 species for which multiple individuals were sequenced for both rbcL and matK to provide a measure of relative discrimination. Species represented by single sequences were included in the analyses to serve as sources of discrimination failure (decoys). The four approaches for measuring discrimination success, barcode gap (pairwise and multiple alignment), monophyletic groups in NJ trees and BLASTn, provided broadly similar results (Fig. 3 & Dataset S3). Relative discrimination across all four methods was highest using a combined rbcL and matK matrix, the different methods provided a range of 69.4–74.9% discrimination at the species level and 98.6–99.8% discrimination to genus. matK performed well on its own, with 68.7–74.1% of species and 98.0–99.1% of genera discriminated. This compares with 55.8–60.9% of species discriminated with rbcL and 94.3–97.2% of genera.


DNA barcoding the native flowering plants and conifers of Wales.

de Vere N, Rich TC, Ford CR, Trinder SA, Long C, Moore CW, Satterthwaite D, Davies H, Allainguillaume J, Ronca S, Tatarinova T, Garbett H, Walker K, Wilkinson MJ - PLoS ONE (2012)

Ability of the DNA barcode markers rbcL and matK to discriminate the Welsh flora.Discrimination (%) at species, genus and family level for rbcL, matK and both markers combined using monophyletic groups in Neighbour-Joining trees (Tree), BLASTn searches (BLAST) and barcode gap analysis using pairwise (Barcode gap pairwise) and multiple alignments (Barcode gap multiple). Species level discrimination for monophyletic groups in Neighbour-Joining trees is divided into bootstrap support values of ‘any’, >50% and >70% based on 1000 bootstrap replicates. Discrimination uses 808 species for which multiple individuals were DNA barcoded for both rbcL and matK. Species with single sequences were included in the analyses as sources of discrimination failure. For a complete list of which species can be discriminated using the different methods see Dataset S3.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3368937&req=5

pone-0037945-g003: Ability of the DNA barcode markers rbcL and matK to discriminate the Welsh flora.Discrimination (%) at species, genus and family level for rbcL, matK and both markers combined using monophyletic groups in Neighbour-Joining trees (Tree), BLASTn searches (BLAST) and barcode gap analysis using pairwise (Barcode gap pairwise) and multiple alignments (Barcode gap multiple). Species level discrimination for monophyletic groups in Neighbour-Joining trees is divided into bootstrap support values of ‘any’, >50% and >70% based on 1000 bootstrap replicates. Discrimination uses 808 species for which multiple individuals were DNA barcoded for both rbcL and matK. Species with single sequences were included in the analyses as sources of discrimination failure. For a complete list of which species can be discriminated using the different methods see Dataset S3.
Mentions: In order to compare across markers and methods of discrimination, we used a dataset of 808 species for which multiple individuals were sequenced for both rbcL and matK to provide a measure of relative discrimination. Species represented by single sequences were included in the analyses to serve as sources of discrimination failure (decoys). The four approaches for measuring discrimination success, barcode gap (pairwise and multiple alignment), monophyletic groups in NJ trees and BLASTn, provided broadly similar results (Fig. 3 & Dataset S3). Relative discrimination across all four methods was highest using a combined rbcL and matK matrix, the different methods provided a range of 69.4–74.9% discrimination at the species level and 98.6–99.8% discrimination to genus. matK performed well on its own, with 68.7–74.1% of species and 98.0–99.1% of genera discriminated. This compares with 55.8–60.9% of species discriminated with rbcL and 94.3–97.2% of genera.

Bottom Line: We have sampled multiple individuals for each species, resulting in 3304 rbcL and 2419 matK sequences.These approaches yield similar results, providing relative discrimination levels of 69.4 to 74.9% of all species and 98.6 to 99.8% of genera using both markers.Species discrimination can be further improved using spatially explicit sampling.

View Article: PubMed Central - PubMed

Affiliation: National Botanic Garden of Wales, Llanarthne, United Kingdom. natasha.devere@gardenofwales.org.uk

ABSTRACT
We present the first national DNA barcode resource that covers the native flowering plants and conifers for the nation of Wales (1143 species). Using the plant DNA barcode markers rbcL and matK, we have assembled 97.7% coverage for rbcL, 90.2% for matK, and a dual-locus barcode for 89.7% of the native Welsh flora. We have sampled multiple individuals for each species, resulting in 3304 rbcL and 2419 matK sequences. The majority of our samples (85%) are from DNA extracted from herbarium specimens. Recoverability of DNA barcodes is lower using herbarium specimens, compared to freshly collected material, mostly due to lower amplification success, but this is balanced by the increased efficiency of sampling species that have already been collected, identified, and verified by taxonomic experts. The effectiveness of the DNA barcodes for identification (level of discrimination) is assessed using four approaches: the presence of a barcode gap (using pairwise and multiple alignments), formation of monophyletic groups using Neighbour-Joining trees, and sequence similarity in BLASTn searches. These approaches yield similar results, providing relative discrimination levels of 69.4 to 74.9% of all species and 98.6 to 99.8% of genera using both markers. Species discrimination can be further improved using spatially explicit sampling. Mean species discrimination using barcode gap analysis (with a multiple alignment) is 81.6% within 10×10 km squares and 93.3% for 2×2 km squares. Our database of DNA barcodes for Welsh native flowering plants and conifers represents the most complete coverage of any national flora, and offers a valuable platform for a wide range of applications that require accurate species identification.

Show MeSH