deletions | additions       diff --git a/Concepts.tex b/Concepts.tex  index 286da0d..00765d1 100644  --- a/Concepts.tex  +++ b/Concepts.tex 
...
argument for the burst selection function \verb|select_bursts.ES| (see  section~\ref{sec:burstsel}).  \subsection{Introduction to burst search}  \label{sec:burstsearch_intro}  After background estimation, the burst search is the next fundamental step of  the analysis. The core "sliding window" algorithm, proposed by  Eggeling~\textit{et al.} in 1998~\cite{Eggeling_1998}, involves searching for  bursts of photons  in which $m$ consecutive photons are contained within a minimal time period  $\Delta t$. In other words, bursts are portions of the photon stream where the  local rate (computed using $m$ photons) is above a minimal rate chosen as a  threshold. Eggeling did not provide any criteria on how to choose the rate  threshold and the number of photons $m$ and as therefore it has become a common  practice to manually tweak those parameters for each specific measurement.   A more general approach consists in taking into account the background rate of  the specific measurements and in choosing a rate threshold that is $F$ times  larger than the background rate. This approach assures that all the resulting bursts  have a Signal-To-Background ratio (SBR) larger than  $(F-1)$~\cite{Michalet_2012}. A consistent criterion for choosing the threshold is  very important when comparing different measurements with different background  rates, when the background significantly varies during measurements or in  multi-spot measurements where each spot has a different background rate.  A second important aspect of burst search is which photon stream is processed.  In some cases, for instance when identifying FRET populations, one would like to  apply the burst search to all the photons. Other times, when focusing on  donor-only or acceptor only populations, it is better to use only the donor or  acceptor signal. In general one would like to be able to apply the burst search  to an arbitrary selection of photons. In FRETBursts this can be achieved by passing  the appropriate \verb|Ph_sel| object to the burst search method (see  section~\ref{sec:ph_streams} for more info on photon stream definitions).  Finally, Nir~\textit{et al.}~\cite{Nir_2006} proposed a Dual-Channel Burst  Search (DCBS) that can help mitigating artifacts due to  photo-physical effects such as photo-blinking. In this case a search is performed  independently on two photon streams and bursts are marked only when both photon  streams exhibit a rate higher than the threshold,   implementing a kind of an AND-gate logic.   Usually, the term DCBS refers to a burst search where the two photon streams  are (1) all the photons   during donor excitation (\verb|Ph_sel(Dex='DAem')|) and (2) acceptor channel photons  during acceptor   excitation (\verb|Ph_sel(Aex='Aem')|).  After each burst search it is important to select  bursts according to their number of photons (burst size). In the most  basic form, this selection can be performed during burst search by discarding  bursts with size smaller than a threshold $L$, as originally proposed by   Eggeling~\textit{et al.}~\cite{Eggeling_1998}.   This method, however, neglects the effect  of background and gamma factor on the burst size and can lead to a selection  bias of certain channels and/or sub-populations.   For this reason we advocate performing a burst size selection after background  correction and taking into account the gamma factor, as illustrated in  section~\ref{sec:burstsel}.  \subsection{Burst weights}  As will be shown in the following sections, the burst search returns bursts of different sizes (i.e. number of photons).   Bursts with large size (that indeed contain most of the information)