deletions | additions       diff --git a/Results & Discussion.tex b/Results & Discussion.tex  index d8679ca..9abcb43 100644  --- a/Results & Discussion.tex  +++ b/Results & Discussion.tex 
...
Proteobacteria represent 47\% of all cellulose responders identified. Of those, Cellvibrio (named for it's cellulose degrading abilities) accounted for 5\% of all Proteobacterial responders detected. Details about max density shift (need to talk to Chuck because this data is not easy to acquire). Other prominent cellulose degrading Proteobacteria, Stenotrophomonas and Devosia have previous been demonstrated in degrading cellulose \cite{Trujillo_Cabrera_2012, Verastegui_2014}. Brevundimonas has not previously been identified as a cellulose degrader, but has been show to degrade cellouronic acid, an oxidized form of cellulose \cite{Tavernier_2008}.  \textit{This subsection needs a name.} \textit{Responder Characteristics.}  There were 6 shared responders of all responders (n = 72) between xylose and cellulose treatments. Stenotrophomonas, Planctomyces, Rhizobiaceae, Comamonadaceae, Cellvibrio. Two of these shared responders corresponded in time between the two treatments (supplemental table); Cellvibrio (day 3) and Planctomyces (day 14). The kernel density estimate (KDE) of buoyant density shifts resulting from \textsuperscript{13}C-assimilation reveal that cellulose responders exhibit a significantly (wilcox: p\textless...) greater buoyant density shift than xylose responders (Figure 3A). First, a density profile for each responder is generated for the experimental and control treatment at each of the sampling time points using relative abundances from sequence libraries (Fig Sx). Then the center of mass is measured for the density profile of a responder in the control and respective experimental treatment for a single time point. A density shift is then calculated by differentially calculating the center of mass for the control from the experimental. Finally, this is repeated for all responders at each time point for both the \textsuperscript{13}C-cellulose and \textsuperscript{13}C-xylose treatments. Each KDE curve represents the collection of density shifts calculated for all responders at each time point within a treatment (Figure 3A). The ‘difference in center of mass’ is a function of both the density shift for individual DNA molecules and the heterogeneity of labeling for the population of cells within that OTU. As such, we would not expect to observe the maximum density shift (0.04gmL\textsuperscript{-1}) measured in pure cultures with 100\% \textsuperscript{13}C-labeling of DNA (cite). We observe xylose utilizers having a smaller density shift (\textless0.02 gmL\textsuperscript{-1}) than cellulose utilizers (0.005-0.03 gmL\textsuperscript{-1}), with few exceptions. This suggests greater substrate specificity among cellulose degraders than xylose degraders. Partial \textsuperscript{13}C-labeling (\textless 0.04gmL\textsuperscript{-1} density shift) could be a result of various trophic strategies such as (1) assimilation of C from multiple substrates (both \textsuperscript{12}C and \textsuperscript{13}C in this instance) or (2) \textsuperscript{13}C-label dilution as it cascades through trophic levels via consumption of \textsuperscript{13}C-labeled organisms or waste products from organisms that are metabolizing the \textsuperscript{13}C-substrate.