Committee Chair

Shaw, Joey

Committee Member

Carver, Ethan; Chatzimanolis, Stylianos


Dept. of Biological and Environmental Sciences


College of Arts and Sciences


University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)


DNA barcoding is the use of short standardized regions of DNA to identify unknown specimens to species. Currently, the zoological community has agreed that cytochrome oxidase I subunit 1 (COI), a mitochondrial gene region, will serve as the barcode region for all animal taxa. Due to oftentimes complicated evolutionary histories of plants, the plant barcoding community has had a much harder time agreeing on a gene region or regions that should be used to barcode the various land plant lineages. This is in large part due to poor reproductive barriers, which allows for chloroplast sharing between closely related species. In the summer of 2009, the Consortium for the Barcoding of Life‘s Plant Working Group (CBOL, PWG) announced that portions of two coding chloroplast gene regions (cpDNA), matK and rbcL, would serve as the DNA barcode for all land plants. This recommendation was accompanied by CBOL‘s call for continued testing of these two regions, along with other potential gene regions that may prove to be more effective barcode regions, such as noncoding cpDNA regions like trnH-psbA. Originally, this project was focused on the utility of three noncoding cpDNA regions (trnH-psbA, trnL-trnL-trnF, and trnS-trnG-trnG) at identifying species from the genus Prunus L. Upon the announcement by CBOL, additional sequence data was generated for matK and rbcL using the same Prunus taxa to determine how well these two regions would delimit species compared to the three noncoding cpDNA regions. In addition to this, sequence data for matK and rbcL were generated for 27 angiosperm taxa and compared to 34 previously tested noncoding chloroplast gene regions to determine their relative genetic variability. This broader study enabled me to directly compare the genetic variability of these two coding regions to that of noncoding regions. My results for the broader study demonstrate that matK and rbcL contain less genetic variability than noncoding regions. Based on the number of potentially informative characters (PIC), matK was the 25th most variable region and rbcL was the 34th most variable region out of 36 regions tested. This low level of genetic variability in these two regions may make it difficult to identify closely related species. I recommend further study of the 34 previously tested noncoding cpDNA regions to determine their respective utility as plant DNA barcodes. For the direct species identification tests using Prunus, I found that no region alone or in combination was able to discriminate > 50% of species, and noncoding cpDNA regions typically outperformed the Consortium for the Barcode of Life‘s combination of matK+rbcL.


My research was funded by the UC Foundation Faculty Research Grant for J. Shaw, the UTC Graduate School and Provost Award, and a grant from the National Science Foundation.


M. S.; A thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Master of Science.




DNA; DNA data banks


DNA barcoding; Prunus laurocerasus


Biology | Environmental Sciences | Life Sciences | Physical Sciences and Mathematics

Document Type

Masters theses




xi, 96 leaves