Genomic studies of low-biomass environments is often limited by the amount of DNA available, and one solution has been to use a whole genome amplification technique that uses phi29 DNA polymerase known as multiple displacement amplification (MDA).
When used in single cell genomic studies, one noted drawback of the procedure has been amplification bias that impacts quantitative analyses. In a paper published in the December 2010 issue of Nature Methods, former DOE JGI researcher Phil Hugenholtz teamed with Susan Yilmaz and Martin Allgaier to find out if similar amplification biases were detected in studying microbial communities.
|Overview of the multiple displacement amplification process. (Image by Quanlim via Wikimedia Commons)
MDA was first performed on environmental DNA samples extracted from garden compost, activated sludge and termite hindgut, and biases were assessed using high-throughput small-subunit rRNA gene amplicon pyrosequencing. The team found that anywhere from three and 28 percent of the taxa detected in the samples were skewed, with relative abundance numbers unpredictably increased or reduced.
“We find that not only does MDA introduce significant and reproducible biases in population relative abundance, different MDA protocols introduce opposing skews in many populations making it impossible to predict introduced bias for a given microbial community. This essentially precludes quantitative comparative analyses of metagenomic datasets using MDA,” noted Hugenholtz.