Investigations of the Accuracy of Partial Genome Sequencing and Restriction Fragment Analysis in Determination of Genetic Relationships: A Computer Simulation Study

A computer simulation approach has been used to evaluate the accuracy of partial genome sequencing and restriction fragment analysis in determination of genetic relationships, as represented in phylogenetic trees. Three partial genome DNA sequences with different lengths (200, 600, 1000 bps) and three analytic algorithms for restriction fragment data (the Dice coefficient, the Nei and Li distance, and the Weigel and Scherba distance) were considered. The effects of genetic factors (genome length, overall nucleotide substitution rate, and extent of rate variation among sites) and model tree parameters (extent of rate variation among lineages, tree imbalance, and tree stemminess) on accuracy were also evaluated. Thirty-two model trees, each with different values of the three tree parameters, were selected. For each model tree, genomic DNA sequences were simulated, with resultant variation in the three genetic factors. The generated extant DNA sequences were analyzed using partial genome sequencing and restriction fragment analysis, and the phylogenetic relationships among the DNA sequences were reconstructed. These phylogenetic trees were then compared with the model trees, and accuracy was evaluated for two different aspects: similarity of topology and similarity of branch lengths. The analysis indicated that partial genome sequencing generally was more accurate in phylogenetic reconstruction than restriction fragment analysis. However, many factors modified the relative accuracy of these two techniques. Longer partial genome sequences produced more accurate results. The three analytic algorithms for restriction fragment data had different accuracy, with the Dice coefficient and the Nei and Li distance differing little, with both more accurate than the Weigel and Scherba distance. The associations of genetic factors and model tree parameters with the accuracy of phylogenetic reconstruction were complicated and varied among techniques, analytic conditions, and the measure of accuracy used. Based on findings from this study, some recommendations regarding the utility of the two molecular techniques and the analytic algorithms, and the selection of gene or region for sequence analysis were provided.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Veterinary Pathobiology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Qiao, Baozhen

Contributors dc:contributor

• Weigel, Ronald M.

• Computer Science

Language dc:language

eng

Identifier

(MiAaPQ)AAI3086161

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/87621