Note to users. If you're seeing this message, it means that your browser cannot find this page's style/presentation instructions -- or possibly that you are using a browser that does not support current Web standards. Find out more about why this message is appearing, and what you can do to make your experience of our site the best it can be.

Science 18 August 2006:
Vol. 313. no. 5789, p. 918
DOI: 10.1126/science.1126593
Prev | Table of Contents | Next

Technical Comments

Comment on "Computational Improvements Reveal Great Bacterial Diversity and High Metal Toxicity in Soil"

John Bunge1*, Slava S. Epstein2 and Daniel G. Peterson3

1 Department of Statistical Science, Cornell University, Ithaca, NY 14853, USA.
2 Department of Biology and Marine Science Center, Northeastern University, Boston, MA 02115, USA.
3 Department of Plant and Soil Sciences, Mississippi State University, Mississippi State, MS 39762, USA.


Figure 1 Fig. 1. Cot curves fitted to noncontaminated soil data (2) by nonlinear regression. (A) Mixture-of-three-point-masses species-abundance model, used in equation 11 in (15), with parameters estimated by nonlinear least-squares regression, yields function shown by solid line; data points are overlaid. (B) Fitted curve extended to complete (100%) reassociation. Nonconstant curvature is due to the mixture of Cot curves with varying reassociation rates. Extension of estimated curves far beyond available data is statistically inadvisable. [View Larger Version of this Image (7K GIF file)]
 

Figure 2 Fig. 2. Equating {Delta}h with DNA reassociation in complex samples can produce misleading results. (A) DNA extracted from a soil sample represents numerous bacterial species/strains as shown, with 90% of the DNA contributed by several strains of Species G. For simplicity, assume that different species share no notable sequence homology but that DNA from strains of the same species can form duplexes during reassociation (with occasional base mismatches due to modest sequence divergence). (B) A hydroxyapatite chromatography–based Cot curve of the soil DNA extract would show rapid reassociation of Species G DNA (red portion of curve) compared with DNA of other species (blue portion). Although the Species G genome may contain little repetitive sequence, its relative abundance in the DNA extract would cause it to reassociate at least 100 times as fast as DNA of any other species. The gap in relative sequence redundancy between Species G and DNA sequences of other species would result in a flat region of the curve where there would be no notable DNA reassociation (black portion). (C) Cot curve prepared from the same soil extract, in which {Delta}h data are used to estimate DNA reassociation. For simplicity, assume that {Delta}h from complete native double-stranded DNA to complete denaturation accounts for a 27% change in absorbance (9) and that repetitive DNA (here, Species G DNA duplexes) exhibit half the {Delta}h of native DNA, as is typical of eukaryotic repeats (9, 10). As a result of its relatively low hypochromicity, reassociation of Species G DNA will occupy only 12% of the abscissa (0.27 x 0.5 x 0.9 = 0.12). At high Cot values (e.g., 104 M·s), reassociation of soil extract DNA will appear to be far from completion (i.e., 100% hypochromicity), when in reality it may have finished reassociating. (D) Reassociation of Species G DNA at relatively low Cot coupled with its reduced {Delta}h may cause some researchers to discount its renaturation as a "collapse" hypochromicity effect; see (16) for definition. Consequently, they may entirely omit it from their Cot curve, as shown. Extrapolation of the curve to 100% hypochromicity (dotted blue line) would amplify the error. [View Larger Version of this Image (21K GIF file)]
 

To Advertise     Find Products

Science. ISSN 0036-8075 (print), 1095-9203 (online)