Heavy Meson Analysis

Particles like the $\omega$ , $\eta$ , $\eta'$ and $D_0$ decay in a similar way into two oppositly charged pions and a neutral decay meson, that can either be a neutral pion, or in the case of the $\eta'$ an $\eta$ meson.

These analyses share a common analysis task: AliAnalysisTaskNeutralMesonToPiPlPiMiPiNeutralMeson.cxx and can use the same script of the afterburner with different settings for the different mesons and involved NDM.

For the 7 and 13 TeV analysis of the $\omega$ meson, this was done using a post-processing chain explained on the bottom of this page. Since then, a second version (V2) was created, which is explained in this section:

The entire post-processing from the signal extraction up to the production of the final results is managed be the start_FullPiPlPiMiNDMAnalysis.sh macro. Different parameters can be used to trigger different parts of the analysis chain:

-e triggers the extraction of the raw yield in data in ExtactSignalPiPlPiMiPiZeroV2.C
-f triggers the full analysis, which includes the signal extraction in both data and MC using ExtactSignalPiPlPiMiNDMV2.C, as well as the correction of the yield using CorrectSignalPiPlPiMiNDMV2.C
-p triggers the plotting of all monitoring plots using PiPlPiMiNDMMonitoring.C
-c triggers a cutstudy, comparing the extracted histograms for the different cutsettings
-s triggers the calculation of the systematic uncertainties, however, this script will have to be very much personalized for each analysis
-b triggers the combination of the spectra from the different NDM reconstruction methods using the blue method
-r triggers the calculation and plotting of the final results
-d activated the debug mode, in which only one cutstring will be run for code debugging
-0 switches on the analysis of the $\pi^{0}$
-h or simply no argument at all will display a short explaination and a list of the possible arguments

Analysis settings and data paths are specified in CutSelection.csv, included as a template in your Afterburner package. Every row of this table corresponds to one analysis/variation, for which the signal extraction and correction is performed. The systematics and combination are then done based on the entire table.

The first column states, whether a given row should be analyzed or simply ignored. To have the analysis run over a given row simply put an x or similar as the first entry. The second column is used to cluster different cuts into groups of systematic uncertainties, but this is only needed in for the calculation of the systematic uncertainties, so we will ignore this column for now. The next fields are pretty self explainatory: The analysed energy, meson, data and MC period, as well as the paths to the LEGO train output. Then you have to give the Trainconfig and enter a cutnumber included in the file. The next field asks for a ExtractionCutnumber, which is a five digit number, that similar to the established cutstring sets different parameters for the signal extraction. For details on this, please have a look at the PiPlPiMiNDM.h file, in which the different possible parameters are then set, based on the ExtractionCutnumber. The default setting is 00000. The next field asks for the PhotonMode, which you can take from the following table:

mode

reconstruction technique

Finally you can enter a description used for some plotting and comparison purposed (cutstudies), and the $p_{T}$ range, in which the signal is to be extracted. The binning has to be set per energy and meson in ExtractSignalBinning.h.

After configuring the analysis parameters and data paths in the CutSelection.csv for you analysis, all you have to do to get your corrected cross-sections and everything else is:

  start_FullPiPlPiMiPiZeroAnalysis.sh -fp

The following files are part of the NDM post-processing chain:

ExtractSignalPiPlPiMiNMDV2.C: Transforms the train output into raw meson yields - applied to both data and MC
CorrectSignalPiPlPiMiNMDV2.C: Combines the raw data and MC yields to create a corrected yield and cross-section
PiPlPiMiNDM.h: General heavy meson header file including the mass ranges to be set for each meson and energy
PiPlPiMiNDMMonitoring.C: Plotting of the signal extraction and correction (inv. mass bins, efficiency, S/B, ...)
CutStudiesPiPlPiMiPiZero.C: Compares the output from the signal extraction and correction script for all cutstrings in the CutSelection.csv file
SystematicsPiPlPiMiPiZero.C: Calculates the systematic uncertainty for the 5TeV pp and pPb omega analysis - here, every analysis will need its own file
CombineSignalPiPlPiMiPiZero.C: Combines the corrected yields/cross sections from different photon reconstruction methods using the BLUE method
ProduceFinalResultsPiPlPiMiNDM.C: Plots the cross-sections, $R_{pA}$ and NDM/ $\pi^{0}$ ratio

Previous heavy meson analysis workflow:

Because the $\omega$ and the $\eta$ can both decay into $\pi^+\pi^-\pi^0$ , the analysis in this decay channel is organized in a single analysis task: AliAnalysisTaskNeutralMesonToPiPlPiMiPiZero.cxx.

Analogue to the neutral pion analysis, there is a single macro that starts the first afterburners needed to process the output files of the task. The start_FullOmegaMesonAnalysis.sh macro -- even though the name might be misleading -- takes care of the signal extraction (using ExtactSignalPiPlPiMiPiZero.cxx) and its correction (using CorrectSignalPiPlPiMiPiZero.cxx) for $\omega\rightarrow\pi^+\pi^-\pi^0$ and $\eta\rightarrow\pi^+\pi^-\pi^0$ .

  start_FullOmegaAnalysis.sh [-$OPTION] $Data-file.root [$MC-file.root] eps

In case either the data- or MC-file is missing, you can just pass the macro a dummy string. E.g the command

bash start_FullOmegaAnalysis.sh /Path/To/datafile.root bla eps

will run the signal extraction macro two times -- once for the $\omega$ and once for the $\eta$ -- but won't run the correction macro afterwards, because it detected that you didn't provide it with a valid MC file.

There are several options available for this macro (which are still work in progress) that let you e.g. only analyse the $\omega$ (--omegaOnly) or the $\eta$ (--etaOnly). For a full list of available options please use the -h flag.

If you start the start_FullOmegaAnalysis.sh macro by hand, you will be asked about several things needed for the signal extraction and correction. The questions are similar to the ones asked e.g. by start_FullMesonAnalysis_TaskV3.sh, however there are some differences, so we will go through the questions one-by-one in the following chapter.

start_FullOmegaAnalysis.sh

First we start the macro using (e.g.):

bash start_FullOmegaMesonAnalysis.sh /referenceDirectory/Legotrain-vAN-20171122-7TeV-omegaAnalysis/LHC10_GammaConvNeutralMesonPiPlPiMiPiZero_0_29.root /referenceDirectory/Legotrain-vAN-20171122-7TeV-omegaAnalysis/LHC14j4_GammaConvNeutralMesonPiPlPiMiPiZero_0_29.root eps

First, the macro will print the full path of the input files you specified. This comes in handy to track down if you got something wrong while using e.g. relative paths. The first thing you will be asked is to specify the method (mode) used to reconstruct the decay-photons of the $\pi^0$ :

The data file specified is /data/alice/pp7TeV/Legotrain-vAN-20171122-7TeV-omegaAnalysis/LHC10_GammaConvNeutralMesonPiPlPiMiPiZero_0_29.root
The MC file specified is /data/alice/pp7TeV/Legotrain-vAN-20171122-7TeV-omegaAnalysis/LHC14j4_GammaConvNeutralMesonPiPlPiMiPiZero_0_29.root
Which mode are you running? 40 (PCM-PCM *), 41 (PCM-EMCAL *), 42 (PCM-PHOS *), 43 (PCM-DCAL), 44 (EMCAL-EMCAL *), 45 (PHOS-PHOS*), 46 (DCAL-DCAL), 47 (PCM-DALITZ), 48 (EMCAL-DALITZ), 49 (PHOS-DALITZ), 50 (DCAL-DALITZ)

As you may have noticed, the mode numbers used for this analyisis differ from the ones used for the $\gamma\gamma$ -analysis. This was implimented to avoid any conflicts between the two analysis (e.g. between $\eta\rightarrow\gamma\gamma$ and $\eta\rightarrow\pi^+\pi^-\pi^0$ ). The modes currently implemented are marked with a '*'. For the sake of completeness, you can find a nicely formated table of the numbering scheme used in this analysis below:

mode

reconstruction technique

This information will later be used by the signal extraction macro to check if your choice is actually consistent with the input files you provided. After you entered the desired mode, you will be asked

Do you want to take an already exitsting CutSelection.log-file. Yes/No

If you select no, the macro will write all the cut strings found in the input file in a file called CutSelection.log sperated by [ENTER] (using the MakeCutLog.C macro) and then run the signal extraction (and correction) for each cut string in that file. If you select yes, the macro will search for an existing CutSelection.log file in the current working directory and run the afterburners for all cut strings contained in that file. This option is therefore useful if you just want to run the afterburners on a particular set of cut strings.

Let's assume we want to run the afterburners on all the cuts contained in the input file, so we answer with "no". The output should look something like this:

-> found TopDir: GammaConvNeutralMesonPiPlPiMiPiZero
0_00000113_00200009327000008250400000_0103503800000000_302010708_0153503000000000
0_00000113_00200009327000008250400000_0103503800000000_302010708_0b53503000000000
0_00000113_00200009327000008250400000_0103503800000000_302010708_0c53503000000000
0_00000113_00200009327000008250400000_0103503800000000_302010708_0d53503000000000
Which collision system do you want to process? 8TeV (pp@8TeV), 7TeV (pp@7TeV), 13TeV (pp@7TeV, 900GeV (pp@900GeV), 2.76TeV (pp@2.76TeV), PbPb_2.76TeV (PbPb@2.76TeV), pPb_5.023TeV (pPb@5.023TeV)

Make sure that the macro detected all the cuts correctly and proceed by entering the collision system you are analysing. In this example, we will choose pp@7Tev.

The collision system has been selected to be 7TeV.
How many p_T bins do you want to use for Omega? 36(7gev), 37(8gev), 38(10gev), 39(12gev), 40 (16gev), 41 (20gev), 42 (25gev)

After the collision system was selected, you have to specify how many $p_T$ -bins should be analysed. This question is a bit misleading, because you are actually specifying up to what index in your $p_T$ -bin array stored in ExtractSignalBinning.h you want to do the analysis. The total number of analysed bins would then be $N_{\text{StartBin}}-N_{\text{EndBin}}$ . Important: Make sure that the $p_T$ -bin choosen here is not actually greater than the length of the array containing your binning for this analysis (see ExtractSignalBinning.h) ! Also: At the moment only the use of same number of $p_T$ bins for $\omega$ and $\eta$ are supported.

You have chosen  14  pt bins for Omega
mode has been chosen: 40 
I went into standard modes
Which fit do you want to do? CrystalBall or gaussian convoluted with an exponential function? CrystalBall/Gaussian?
Gaussian
Gaussian chosen ...
Please check that you really want to process all cuts, otherwise change the CutSelection.log. Remember at first all gamma cutstudies will be carried out. Make sure that the standard cut is the first in the file. Continue? Yes/No?
Yes

After choosing the desired fitting function for the $\omega$ / $\eta$ -peak (using gaussian is recommended, no testing was done for crystal ball so far), you will be asked if you are sure to process all cuts. After confirming this final question, the afterburners should go to work!

ExtractSignalPiPlPiMiPiZero.C

This macro is used to extract the invariant mass distribution of the meson signal for each analyzed $p_T$ -bin and its usage and tasks are almost identical to ExtractSignalV2.C. You can start the macro using:

root -b -x- q -l 'TaskV1/ExtractSignalPiPlPiMiPiZero.C+("$MESONAME","/path/to/input.root","$CUTNUMBER","$SUFFIX","$MCOPTION","$ENERGY","Gaussian","","","",$NPTBINS,$OPTIONADDSIG,$MODE)'

Example usage (Data):

root -b -x- q -l 'TaskV1/ExtractSignalPiPlPiMiPiZero.C+("Omega","/data/alice/pp7TeV/Legotrain-vAN-20171122-7TeV-omegaAnalysis/LHC10_GammaConvNeutralMesonPiPlPiMiPiZero_0_29.root","0_00000113_00200009327000008250400000_0103503800000000_302010708_0d53503000000000","pdf","kFALSE","7TeV","Gaussian","","","",14,kFALSE,40)'

Example usage (MC):

root -b -x- q -l  'TaskV1/ExtractSignalPiPlPiMiPiZero.C+("Omega","/data/alice/pp7TeV/Legotrain-vAN-20171122-7TeV-omegaAnalysis/LHC14j4_GammaConvNeutralMesonPiPlPiMiPiZero_0_29.root","0_00000113_00200009327000008250400000_0103503800000000_302010708_0d53503000000000","pdf","kTRUE","7TeV","Gaussian","","","",14,kFALSE,40)'

Further Processing

The following macros that allow further processing of the extracted signals have been implemented for $\omega$ / $\eta$ analysis so far:

PreviousNeutral Pion and Eta Analysis NextMerged Cluster Analysis

Last updated 2 months ago