Microbial Evolution & Genomics | Bioinformatics
Estimating population diversity with unreliable low frequency counts
John Bunge*, Dankmar Böhning, Heather K Allen, James A Foster
*Corresponding author: John Bunge
Department of Statistical Science, Cornell University, Ithaca, NY, USA
F1000 Posters 2012, 3: 56 (slide presentation) [English]
Slide Presentation [191.83 KB] | Resulting articles
Presented at
Pacific Symposium on Biocomputing (PSB) 2012,
3 - 7 Jan 2012, P000
When estimating the number of species in a population, often the sample counts of rare species are questionable. This is especially true when the “species” are classifications derived from clustering of sequences produced by next-generation sequencers.
We find that it is possible to statistically discount the seemingly dubious low-frequency sample counts, ex post facto, and thereby adjust the population diversity estimate. However, such procedures depend strongly on their underlying assumptions; it is always preferable to correct the data at the source.
Comparison of statistical discounting with source-data correction will disclose whether the statistical procedures are reasonable, or overly reductive.
No relevant conflicts of interest declared.
National Science Foundation (NSF), DEB-08-16638
Please note that most posters on this site present work that is preliminary in nature and has not been peer reviewed.
This poster is open access subject to the Non-Commercial Creative Commons 3.0 License