tutorial-analysis/_extras/branch_dictionary.md

0001 ---
0002 title: "Branch Dictionary"
0003 ---
0004
0005 Data files from EICrecon can sometimes look a little overwhelming with the large number of branches included. However, for the exercises in this tutorial, we end up needing only a small subset of the branches and the information they contain. A brief overview of some commonly used branches is included below. You might find it useful to refer to these as you're constructing your analysis.
0006
0007 ## Detector Branches
0008
0009 The naming convention of many of the branches is based around the detector they correspond to. For example, take the branch -
0010
0011 ```console
0012 EcalEndcapNClusters
0013 ```
0014
0015 Breaking this down piece by piece, the branch tells us at the start which detector it corresponds to -
0016
0017 ```console
0018 EcalEndcap
0019 ```
0020
0021 ECal -> Electromagnetic calorimeter, Endcap -> The endcap (as opposed to the barrel). The second part tells us a little bit more about which information pertaining to that detector is in this branch -
0022
0023 ```console
0024 ...NClusters
0025 ```
0026
0027 Clusters tells us this is probably something to do with the clusters of hits in this calorimeter. The N is telling is that this is for negatively charged clusters in this detector. If we open this branch, we see the sort of information we might expect to be associated with calorimeter clusters, for example -
0028
0029 ```console
0030 EcalEndcapNClusters.energy
0031 EcalEndcapNClusters.nhits
0032 EcalEndcapNClusters.position.x
0033 ```
0034
0035 So stored in these branches is event by event information on the energy of clusters in this detector, the number of hits per cluster and the x position of those clusters. Remember, each event may have more than one cluster so this branch is actually an array of values for each event.
0036
0037 ## MCParticles
0038
0039 The MCParticles branch contains truth level information on the input events into the simulation and reconstruction. Many simulation studies will involve a comparison to this "true" information. This branch contains event level information for many quantities  -
0040
0041 ```console
0042 MCParticles.PDG
0043 MCParticles.charge
0044 MCParticles.vertex.x
0045 MCParticles.momentum.x
0046 ...
0047 ```
0048
0049 The PDG value tells us what the input particle actually was, we can cross-reference the output of this with the [PDG code](https://pdg.lbl.gov/2007/reviews/montecarlorpp.pdf) to see what our inputs actually were. We can also easily extract the charge, vertex (x component) and momentum (x component) of the particles in our event using the leaves above.
0050
0051 ## ReconstructedParticles
0052
0053 This branch contains information on all of the particles that EICrecon reconstructed in our simulation. This branch includes all reconstructed particles, charged or not. It contains very similar information to the MCParticles branch, but also includes some other key information. For example, since these particles are reconstructed, their PDG is assigned based upon what EICrecon thinks they are. As such, we also get information on quantities such as GoodnessOfPID in this branch.
0054
0055 ## ReconstructedChargedParticles
0056
0057 Similar to reconstructed particles. In this case though, we only get information on reconstructed particles that are charged.
0058
0059 ## ReconstructedParticlesAssociations
0060
0061 This branch contains information on associations betwen MC truth information and reconstructed particles. In this case, this is for all reconstructed particles. We can find the association index matching the index of the chagred particle and compare what we have reconstructed to its actual "true" information. See the examples in [episode 3]({{page.root}}{% link _episodes/03-analysis.md %}) for some example usage of this.
0062
0063 ## ReconstructedChargedParticlesAssociations
0064
0065 Similar to ReconstructedParticlesAssociations, but for reconstructed charged particles.
0066
0067
0068
0069
0070
0071
0072
0073