drkovalskyi / bmm5 Goto Github PK

So far we cover mm, mmK and mmKK final states in genbmm block. @gandreassi what other channels do we need?

spvip is actively used in the analysis, but it's not available directly from the ntuples. One needs to calculate it on the fly. Let's add it to avoid any mistakes.

This issue can lead to mistakes. So let's train it properly. To avoid such confusion in the future, we need to remove cosAlphaXY branch completely.

We need to make the cuts compatible with the 4/3 selection matching the main dimuon trigger.

DsToPhiPi allows to achieve higher purity of kaon selection. Make it also work for pion fakes to complement ks control region.

Cosine of the point angle is very close to one for signal events and most of its discriminating power comes from very short variations in the angle, so it's better to use angle instead of its cosine as it's done for 3D alpha.

The change requires retraining of the MVA, which is time consuming.

We are looking for K_S with a proper topology in each event unpacking all PFCandidates. It's a very time consuming operation. Need to find a way to speed it up.

Options:

• Prescale hadronic K_S. We don't need them all. The down side can be that we may forget about the prescale in the analysis. Probably it's not an issue if we heavily prescale K_S to pipi since the effect will be obvious.
• Use V0 computed in RECO. It may not work for pi+muon signature. Need to check what we have.

We need input to retrain muon MVA. It's easier to do and to apply if all inputs are available in the ntuples

http://t3serv001.mit.edu/~cmsprod/Kraken/agents/reviewd/nanoao/508/BdToKPi_BMuonFilter_SoftQCDnonD_TuneCP5_13TeV-pythia8-evtgen+RunIIAutumn18MiniAOD-102X_upgrade2018_realistic_v15-v2+MINIAODSIM/BC209D08-9205-D74B-A589-C7E4FF9F5ABB.err

Begin processing the 23016th record. Run 1, Event 34787581, LumiSection 918 on stream 0 at 21-Sep-2020 08:47:30.742 EDT
----- Begin Fatal Exception 21-Sep-2020 08:47:31 EDT-----------------------
An exception of category 'StdException' occurred while
[0] Processing Event run: 1 lumi: 918 event: 34787581 stream: 0
[1] Running path 'nanoAOD_step'
[2] Calling method for module BxToMuMuProducer/'BxToMuMuMc'
Exception Message:
A std::exception was thrown.
PerigeeKinematicState::kinematic state passed is not valid!
----- End Fatal Exception -------------------------------------------------
21-Sep-2020 08:47:31 EDT Closed file file:BC209D08-9205-D74B-A589-C7E4FF9F5ABB.root

bkmm_gen_kaon_pdgId is mostly zero on BuToJpsiK_BMuonFilter_SoftQCDnonD_TuneCUEP8M1_13TeV-pythia8-evtgen+RunIISummer16MiniAODv2-PUMoond17_80X_mcRun2_asymptotic_2016_TrancheIV_v6_ext1-v1+MINIAODSIM. When it's not zero it matches to anything but charged kaons

Need to store the soft pion vertexing fit information like probability, number of tracks etc.

We need to be able to match muons to hadrons for the decay in flight. This information is available in recent MC reconstruction cycles. Not sure we will have it for all years. So it may help to add dR matching as well.

We need to be able to get all events for LambdaBToPPi and LambdaBToPK to get the shapes. Currently only B-mesons are kept.

These control samples may provide a better match to mm final state.

For decays like Ks->pipi we have incomplete gen information. Add mass computed from p4 of the gen matched daughters.

For debugging purposes it can be useful to see the full decay tree of the common ancestor.