deletions | additions       diff --git a/To_have_a_fair.tex b/To_have_a_fair.tex  index f142e2f..6020a29 100644  --- a/To_have_a_fair.tex  +++ b/To_have_a_fair.tex 
...
Intriguingly, most sites identified in Foll et al \citep{Foll2014} (Table \ref{tab:H1N14t}) appear in our analysis to only have signal in the first biological replicate. The exception, S6-822, has a strong signal in both replicates and regardless of end point generation analyzed (figure ...). We speculate that the lack of consistent signal/false signal coming from the other sites is caused by the lower average read count per site for the second replicate compare to the first. The population genetic approach used in Foll \citep{Foll2014} appears to be heavily influenced by the first replicate. This leads us to postulate that their result is adversely affected by the large imbalance in counts.  Noticing that the two Two  additional sites identified in Table \ref{tab:H1N15t} show a  more pronounced drug effect after the 12th passage in both replicate  (see Figures \ref{fig:S5-300},\ref{fig:S8-80}). We conclude that IVA had may  not have  fully responded to the treatment by Passage 12 and 12. Thus  our previous analysis, including which included  the last time point collection, is likely  more reliable for the identification of substitution sites.