PPS Framework

This package is used for the Missing Mass Search analysis. The first tool acts on the CMSSW official datasets producing a root NTuple (Skimmer) which has a set of pre-selected CMSSW collections (and physics-computing objects) event by event. After that, another tool should be launched to produce the final analysis file (Analyzer).

Description

Folder Name	Comments
Skimmer	to store CMSSW pre-selected objects in a NTuple (step 1)
Analyzer	produce the final analysis Ttree (step 2)
Plotter	produce automatic plots for your final analysis (using Ttree format)
Utils	helpful scripts and commands

Installation

Download the package from git. It is also important to set your voms-certificate as well as CMSSW (cmsenv) where your plugins are set.

SCRAM_ARCH=slc7_amd64_gcc700; export SCRAM_ARCH
cmsrel CMSSW_10_6_17_patch1
cd CMSSW_10_6_17_patch1/src
cmsenv
git clone https://github.com/dfigueiredo/PPSFramework.git -b CMSSW_10_6_17_patch1_2023
cd PPSFramework/
scram b -j 8
source set_environment.sh
cd working

Once CMSSW and your package is installed, you need to setup everytime you connect in a new terminal:

SCRAM_ARCH=slc7_amd64_gcc700; export SCRAM_ARCH (or setenv SCRAM_ARCH slc7_amd64_gcc700 for tcsh) 
cd CMSSW_10_6_17_patch1/src
cmsenv

Skimmer

Used to create the NTuple (Skimmer) from a CMSSW dataset type format (AOD or miniAOD).

Usage

As an example, edit the file working/RunMissingMassSearches.py and substitute the line:

process.maxEvents = cms.untracked.PSet( input = cms.untracked.int32(-1) )

for

process.maxEvents = cms.untracked.PSet( input = cms.untracked.int32(5000) )

After that, run the following command:

cmsRun RunMissingMassSearches.py year=2017 physics=bjet trigger=True era=D mode=data

Please, pay attention that the options mode (data or mc), year (2017 or 2018) must be set correctly accordingly to the dataset which has been choosen. If you are not using crab, remember to change the input file name to run locally (in general for a test or to debug the code).

Options

Options	Explanation	Comments
physics	(string) muon, electron, bjet, displacedjet, emu, zerobias	to set the correct trigger path for a specific dataset stream
mode	(string) data, mc	If a data sample, it must be set accordingly (same for monte carlo)
era	(string) A, B, C, D	It depends of the year. Used to load PPS proton reconstruction parameters, cutoff or for simulation
year	(string) 2017, 2018	To deal with HLT names and conditions
prescales	(bool) False or True	enable/disable trigger prescales values. Default is False (only for data)
trigger	(bool) False or True	enable/disable the trigger. Default is True.
unmatching	(bool) False or True	True: leptons and jets not associated in the same cone. Default is false
ppstagging	(bool) False or True	True: selecting events with at least one proton per arm. Default is true
debugging	(bool) False or True	True: enabling skimmer debugger. Default is false

Analyzer

The analyzer is used to produce a TTree after some analysis basic cuts and selecting events with one proton per arm. As an example, you can now run the Analyzer to process the NTuple produced by the previous skimmer (RunMissinMassSearches.py):

cd ../Analyzer
./MissingMassNtupleAnalyzer --f ../working/output.root --year 2017 --era D --mode data --physics bjet

Where all the options are following the Skimmer options (same year, era, mode, physics). The output.root file was produced by the Skimmer (RunMissingMassSearches.py). Now, the output file (NTuple_data_bjet.root) can be opened and the analysis variables can be plotted direclty from the Ttree (extra cuts can also be applied).

Compilation

Everytime the source code is changed, you need to recompile it to run (MissingMassNtupleAnalyzer.cc):

make clean
make

Options

Options	Comments
--protonfile	Used to create, from the detector data per physics (muon, electron, bjets), a root file with Multi and Single Xi histograms. This file is used for the --random option
--eventfile	Used to create, from the detector data, a text data format with golded selected events (to be visualized later on Fireworks)
--short	Option to process only the proton selection, skipping the jets or leptons selection
--notrigger	Skip the trigger selection
--random	If there is no proton in one arm, it adds randomly from the Xi data distribution (single and multi). Used for MC cases
--zerobias	Only for zerobias datasets to study background
--noppstagging	Without proton selection (one proton per arm)
--debugging	For debugging the selections. It prints some outputs on the screen
--help	It prints all the options accepted by the code
--mode	mc or data
--era	A,B,C,D,E or F
--physics	electron, muon, bjet or displacedjet (to enable each trigger menu)
--jobid	It adds a string in the end of the output filename. Used in condor submission to avoid file overwriting (string)
--output	It creates an output folder name (string)
--f	Input file name

Important: the options "--protonfile" and "--random" can not be used together! In case you do not have the random file with the protons xi, first run --protonfile to create the root file histogram used for the --random option.

Examples

Pythia8 SD Top

./MissingMassNtupleAnalyzer --f skimmer_output.root --year 2017 --era C --mode mc --physics muon --random --jobid 0

Toy MC (Higgs or Z)

./MissingMassNtupleAnalyzer --f skimmer_output.root --year 2017 --era C --mode mc --physics bjet --jobid 0

Data

./MissingMassNtupleAnalyzer --f skimmer_output.root --year 2017 --era C --mode data --physics displacedjet --jobid 0

Creating Random File

./MissingMassNtupleAnalyzer --f skimmer_output.root --year 2017 --era D --mode data --physics bjet --protonfile

the file RandomProtons_bjet_eraD.root will be created.