Jan. 22 — Scientists around the globe are using Cornell CatchAll software to perform more accurate statistical analyses in fields ranging from microbial ecology to viral metagenomics. Developed by Computer and Information Science professor John Bunge and Cornell Center for Advanced Computing database designers, CatchAll has become the standard package for population diversity analysis.
In the January 2014 publication of Microbial Ecology, scientists report using CatchAll in the analysis of soil contaminated with heavy metals, a pervasive problem in the vicinity of mines and industrial facilities in Southern Poland. Little is known about most bacterial species thriving in such soils and even less about a core bacterial community. Marcin Golebiewski and colleagues at Nicolaus Copernicus University used 16S rDNA pyrosequencing and CatchAll to assess the influence of heavy metals on both bacterial diversity and community structure. It was found that Zinc had the biggest impact in decreasing both diversity and species richness. Understanding biodiversity in polluted areas helps scientists to quantify the detrimental effects of human activity on particular taxonomic groups and to monitor bioremediation efforts.
In another recent study published in Clinical and Vaccine Immunology, Patricia Diaz and colleagues at The University of Connecticut conducted the first comprehensive evaluation of long-term organ transplant immunosuppression on the oral bacterial microbiome. Many organ transplant patients require lifelong immunosuppression in order to prevent transplant rejection. This study found that prednisone had the most significant effect on bacterial diversity and on the colonization of potentially opportunistic pathogens. The researchers used Catchall to calculate the number of observed operational taxonomic units (OTU) and number of estimated OTUs in order to determine species richness.
The latest version of CatchAll was updated in October 2013 and is available for download.
In spring 2014 John Bunge, with Cornell Department of Statistical Sciences Ph.D. student Amy Willis, will release a new software package called breakaway. Written in R, breakaway implements a radical new statistical approach to diversity estimation based on a little-known thread in probability distribution theory, which exploits ratios of sample counts.
A beta version of breakaway is currently available for testing by contacting the authors.
Source: Cornell University Center for Advanced Computing