Download instructions:

Follow the link Instructions

Developer instructions:

Follow the link Developer instructions

Files Governing Analysis Workflow Configuration

All configuration files and parameters which are used to process minitrees are stored in configs/ and data/.

Main configuration file: The file is configs/2015-v1.conf lists the json file, minitree and skim version number, the blinded or unblinded status of minitrees made from collision data, and the location of the .dat file which lists all collision data and MC samples from which minitrees have been made.

The following information is also stored in configs/:

list of datasets which have been skimmed and made into minitrees (configs/datsets.dat)
mass windows (in GeV) used for limit extraction at different WR mass hypotheses. These values are used by analysis.cpp.

Datasets

The datasets which have been processed into minitrees are listed in

configs/datasets.dat

Triggers List of triggers used to filter events at skim level

Utilities

How to estimate cross sections

Once the dataset is entered in the datasets.dat config file putting dashes for cross section, cross section uncetainty, then you can use the script

 ./scripts/getXsec.sh

to update the file with the estimated cross sections Cross sections for SM processes can be found here: https://twiki.cern.ch/twiki/bin/viewauth/CMS/StandardModelCrossSectionsat13TeV

Instructions on how to get the cross sections of a MC sample: https://twiki.cern.ch/twiki/bin/view/CMS/HowToGenXSecAnalyzer

Analysis sequence

Skims

Follow the link Skim

Note: microAOD_cff

In order to make the skims:

scripts/crabSkimJobs/makeAndRunCrabSkimScripts.sh

and then submit the relevant crab tasks

for skim in tmp/skim*.py; do crab submit -c $skim; done

To update the list of datasets with the published skims:

scripts/crabSkimJobs/updateDatasetsWithSkims.sh

MiniTree production

The actual code to produce miniTree works only on LXPLUS at CERN on the batch system using CRAB2.

To load CRAB2 environment in bash:

source /afs/cern.ch/cms/LCG/LCG-2/UI/cms_ui_env.sh
cmsenv
source /afs/cern.ch/cms/ccs/wm/scripts/Crab/crab.sh
voms-proxy-init -voms cms -out $HOME/gpi.out

If the USER is not Shervin, the SCHEDULER should be set to LSF like:

./script/makeLSFTTreeScripts.sh --scheduler=lsf

To check the status of the jobs and merge the jobs:

./script/makeLSFTTreeScripts.sh --scheduler=lsf --check

CAVEAT: the script is set to save the ntuples in Shervin's area on EOS. It has not been yet generalized. So it is not supposed to work for other users in other storage elements on-the-fly. On can also use –scheduler=remoteGlidein to run over the GRID. The storage element is still CERN. LSF has not been tested.

the cfg to be used is

test/runAnalysis_cfg.py

There are four paths defined in runAnalysis_cfg.py:

SignalRegion: require dilepton mass > 200, signal region triggers
FlavourSideband: require dilepton mass > 200, Mu30_Ele30 trigger
LowDiLeptonSideband: requiredilepton mass < 200, signal region triggers
DYtagAndProbe: require tag and probe single mu or double ele triggers

Lepton and jet IDs are applied when these paths are run, and filters on the number of ID'd objects is applied in the SignalRegion, FlavourSideband, and LowDiLeptonSideband (by the dilepton object candidate filters). No such filter is applied in the DYtagAndProbe path.

Once minitrees are made by runAnalysis_cfg.py, analysis.cpp can be used once a grid proxy is created.