Polymerase chain reaction (PCR) has been widely used by microbial ecologists to determine diversity within environmental samples of unidentified species variety. Hypothetically, if a sample containing many different organisms is analyzed by PCR using universal primers (primers that are designed to amplify a fragment of a specific gene from any organism that possesses the gene) genes from every type of organism should be amplified with equal efficiency. This would allow ecologists to establish the species composition of the sample. However, known artifacts that arise during PCR, such as chimeras and heteroduplexes, as well as mutations due to Taq polymerase error may cause an exaggeration of apparent diversity and lead to an overestimation of species diversity within a given sample. Conversely, partiality towards certain sequences in a mixed-template reaction may result in an underestimation of diversity and even a complete overlooking of a genotype. This research attempts to determine the extent to which such PCR bias occurs, which would then answer the question of whether PCR is a suitable diversity analysis technique. Multiple artificial communities composed of differing concentrations of five phytoplankton species (Cholorella, Tetraselmis, Thalassiosira oceanica, Thalassiosira pseudonana, Thalassiosira weissflogii, and an unidentified species from the class Prymnesiophyceae) were made and collected on filters, following the procedures used when actual environmental samples are processed. Following DNA extraction, the samples were subjected to PCR in order to amplify a fragment of the urease (urea amidohydrolase; EC 3.5.1.5) gene. Primer-dimers were removed from the PCR product through a purification procedure. The purified product was ligated into the pGEM-T Easy vector, which was then used to transform competent Escherichia coli. Following numerous minipreps, those with the correct insert (which has an approximate length of 630 base pairs) were sequenced. Sequences were obtained, analyzed, and aligned with known urease sequences. Unfortunately, due to a contamination problem, conclusions have not been reached at the present time. Subsequent research will focus on analyzing community samples believed to be free of the contaminant.

This work has been supported by the NSF OCE-0223135 grant and the Simons Foundation.