Development of Bioinformatic Tools and Procedures to Computationally Support DNA Barcoding

Abstract

This thesis reports the development of several new and/or improved pipelines for investigating nucleotide alignments as part of DNA barcoding and DNA metabarcoding. In addition, the existing barcode regions were investigated to discover if possible mini-barcodes could be used in order to improve identification in a wide range of plant species. The pipelines developed in the project have also been designed to help the general barcode community in a wide range of areas from forensic analysis, food identification, monitoring illicit trade in plants and animals and numerous metabarcoding projects.

From the development of the improved alignment pipeline, users will be able to align the rbcL and matK barcode regions with confidence, with only a minimal amount of manual editing required for matK. The results achieved for the rbcL were roughly the same for the standalone alignment tools tested, and for the matK gene the results were better than the standalone alignment tools but not as good as the manually edited version of the matK alignment.

From the development of the new metabarcoding analysis pipeline, the results showed that by implementing the pipeline the number of reads decreased but the sequences being lost are low quality, by removing these reads the average quality of the reads left increases. The result is significant as the user would like to keep as many reads as possible with also achieving the highest quality reads possible at the same time.

From investigating the barcode regions to find mini-barcodes, it was possible to find regions that could be used as mini-barcode region for both the rbcL and the matK barcode region. To extract these regions primer combinations were designed for the rbcL and for APG-III classification groups for the matK region. One of the rbcL primer combinations achieved 86 % and several of the matK primer combinations achieved above 60%. This result is significant as these primers could now be used along with the metabarcoding approach to extract matK sequences.

Date of Award	31 Mar 2016
Original language	English
Awarding Institution	University of South Wales
Sponsors	National Botanic Garden of Wales
Supervisor	Denis Murphy (Supervisor), Natasha de Vere (Supervisor) & Andrew Ware (Supervisor)