deletions | additions       diff --git a/introduction.tex b/introduction.tex  index 1337168..684cd79 100644  --- a/introduction.tex  +++ b/introduction.tex 
...
While SIP provides a useful tool for characterizing in situ microbial activity the method has notable limitations including: the need to add substrate at concentrations higher than typical in situ \cite{radajewski2000stable}, the potential for partial labeling due to label dilution \cite{radajewski2000stable,Manefield_2002,McDonald_2005}, the potential for cross-feeding and trophic cascades \cite{Morris_2002,Hutchens_2003,Lueders_2003,DeRito_2005,Mahmood_2005,McDonald_2005}, Ziegler et al. 2005), \cite{Morris_2002,Hutchens_2003,Lueders_2003,DeRito_2005,Mahmood_2005,McDonald_2005,Ziegler_2005}  and variation in genome G+C content which causes variation in native DNA buoyant density (Buckley 2007, \cite{Buckley_2007},  Birnie and Rickwood 1978, Holben and Harris 1995, Nusslein and Tiedje 1998). Several strategies have been developed to deal with these issues and each requires the collection and analysis of density gradient fractions in order to determine the degree of isotope incorporation into nucleic acids from particular microbial groups (Manefield et al. 2002a, Manefield et al. 2002b, Lueders et al. 2004a). While nucleic acid SIP has been around for nearly a decade certain limitations of the technique have constrained its use. The most successful applications of this technique have been to identify organisms which mediate processes performed by a narrow set of organisms (ie: methanotrophs, methanogens, syntrophs, etc). The technique has been less applicable to the study of the soil C-cycle because of limitations in its resolving power which are manifest when many different organisms in the community are simultaneously labeled. The use of massively parallel sequencing, however, should effectively overcome this barrier and should greatly expand the ability of nucleic acid SIP to explore complex patterns of C-cycling in microbial communities. Tag sequencing can be used to sequence 16S rRNA gene fragments from many different gradient fractions and multiple gradients on a single machine run obtaining sequencing depth of thousands of sequences per gradient fraction. This will make it possible to follow carbon assimilation simultaneously in hundreds of different taxa.  Our experimental approach employs the addition of a complex soil organic matter simulant (a mixture of model carbon sources and inorganic nutrients common to plant biomass) to soil. The design of the SOM simulant allows for different components to be exchanged with their 13C-labeled alternatives. Soils are then incubated in parallel with addition of unlabeled SOM or different SOM mixtures in which a different individual components are 13C-labeled. Following incubation, DNA is extracted and subjected to density gradient ultracentrifugation and fractionation into gradient fractions of discrete buoyant density. DNA from bacteria in distinct gradient fractions is amplified by PCR using primers with distinct tag sequences. DNA from >200 fractions can then be run simultaneously in one sequencing reaction and deconvoluted based on their primer tags (as described in (Hamady et al. 2008) and below). As a result, by examining nucleic acids at different times it will be possible to track 13C from model substrates into the nucleic acids of different microbial groups, with the density of nucleic acids increasing over time in proportion to the amount of labeled substrate incorporated. This will provide an unprecedented view of how different types of C compounds move through the soil microbial food web over time.