Evaluating sampling strategy for DNA barcoding study of coastal and inland halo-tolerant Poaceae and Chenopodiaceae: A case study for increased sample size.

Peng-Cheng Yao,Hai-Yan Gao,Hong-Qing Li,Ya-Nan Wei,Xiao-Yong Chen,Jian-Hang Zhang,Diego Breviario

doi:10.1371/journal.pone.0185311

Abstract

Environmental conditions in coastal salt marsh habitats have led to the development of specialist genetic adaptations. We evaluated six DNA barcode loci of the 53 species of Poaceae and 15 species of Chenopodiaceae from China's coastal salt marsh area and inland area. Our results indicate that the optimum DNA barcode was ITS for coastal salt-tolerant Poaceae and matK for the Chenopodiaceae. Sampling strategies for ten common species of Poaceae and Chenopodiaceae were analyzed according to optimum barcode. We found that by increasing the number of samples collected from the coastal salt marsh area on the basis of inland samples, the number of haplotypes of Arundinella hirta, Digitaria ciliaris, Eleusine indica, Imperata cylindrica, Setaria viridis, and Chenopodium glaucum increased, with a principal coordinate plot clearly showing increased distribution points. The results of a Mann-Whitney test showed that for Digitaria ciliaris, Eleusine indica, Imperata cylindrica, and Setaria viridis, the distribution of intraspecific genetic distances was significantly different when samples from the coastal salt marsh area were included (P < 0.01). These results suggest that increasing the sample size in specialist habitats can improve measurements of intraspecific genetic diversity, and will have a positive effect on the application of the DNA barcodes in widely distributed species. The results of random sampling showed that when sample size reached 11 for Chloris virgata, Chenopodium glaucum, and Dysphania ambrosioides, 13 for Setaria viridis, and 15 for Eleusine indica, Imperata cylindrica and Chenopodium album, average intraspecific distance tended to reach stability. These results indicate that the sample size for DNA barcode of globally distributed species should be increased to 11–15.

Highlights

Since 2003, the concept of DNA barcoding has attracted the attention of botanical scientists from all over the world [1; 2; 3; 4; 5; 6; 7]
We found that by increasing the number of samples collected from the coastal salt marsh area on the basis of inland samples, the number of haplotypes of Arundinella hirta, Digitaria ciliaris, Eleusine indica, Imperata cylindrica, Setaria viridis, and Chenopodium glaucum increased, with a principal coordinate plot clearly showing increased distribution points
The results of a Mann-Whitney test showed that for Digitaria ciliaris, Eleusine indica, Imperata cylindrica, and Setaria viridis, the distribution of intraspecific genetic distances was significantly different when samples from the coastal salt marsh area were included (P < 0.01). These results suggest that increasing the sample size in specialist habitats can improve measurements of intraspecific genetic diversity, and will have a positive effect on the application of the DNA barcodes in widely distributed species

Summary

Introduction

Since 2003, the concept of DNA barcoding has attracted the attention of botanical scientists from all over the world [1; 2; 3; 4; 5; 6; 7]. ITS and trnH-psbA were recommended as barcodes for plants [9; 10]. The use of these four loci as plant DNA barcodes has become widely accepted. By comparing intraspecific genetic distances in different cases when selecting 2, 5 and 10 individual samples, they found that the average coalescent depth increased as sample size increased (from 0.0049 to 0.0057 and to 0.0070). They recommend a sample size of 5–10 individuals for DNA barcoding [21]. In order to optimize the accuracy of DNA barcode evaluation, the appropriate number of samples remains to be explored

Methods

Results

Conclusion