LLPtriggertiming

Studies for LLP triggers using timing in low-HT events

Installation

Specify '--recurse-submodules' when cloning, or remember to init+update the submodules Delphes should show the tag '3.4.2' and Pythia 'pythia8235'

cd Pythia
make -j9
cd ../Delphes
patch -p1 < ../Delphes.patch
make HAS_PYTHIA8=true  -j9
cd ..

For local runs of Delphes source setup.sh, for all the analysis plots source setup_analysis.sh

Generate minbias

Generate individual events to overlay

cd run_condor
condor_submit condorSubmit.sub #send 1000 jobs x 6000 events
# wait for jobs
cd ../Delphes
./root2pileup ../LLPtrigger_samples/Minbias.pileup ../LLPtrigger_samples/Minbias.10220229*
#at this point can remove the individual Minbias files
cd ../run_condor
#make sure the delphes card points to the correct Minbias.pileup file within Delphes/cards/delphes_card_ATLAS_PileUp.tcl
condor_submit condorSubmitWithPU.sub #send 1000 jobs x 1000 events
hadd MinbiasWithPU55.root MinbiasWithPU55.10220303.*
#remove the individual files

Check trigger rates

The minbias+PU file can be used to verify the trigger rates. The single jet rates are prety decent but 4J15 is clearly off.

cd Analysis
python plot_samples_withPU.py

Shower existing LHE files

This is not strictly needed, and not fully correct since I don't do the proper Pythia matching. It's only useful to have a sample with higher pT spectrum to derive later on the truth2reco calibration

cd Delphes
#check the LHE file within examples/Pythia8/configLHE.cmnd
./DelphesPythia8 cards/delphes_card_ATLAS_PileUp.tcl examples/Pythia8/configLHE.cmnd ../LLPtrigger_samples/n3n4_1TeVsquark_150gev_showerLHE_withPU55.20k.root
#set showered file as input for
python truth_to_reco.py

The output plots are the truth-reco calibration and relative resolution

Derive calibrations

cd Analysis
python truth_to_reco.py
#check the plot h2d.png, should show something like a -4 GeV offset between truth and reco
python reco_to_l1.py
#check the plot L1Jfit.png, should show a good fit to the turn-on curves
#the four fitted parameters are the calibration constants (reco pT to L1 pT) and the values of the smearing function
#those should be ported to plot_daod.py and convertL1Jet.C

#Analysis plots The analysis works on DAOD_TRUTH3 for any signal sample. Select the signal in the map at the top of the file

cd Analysis
python plot_daod.py