###### tags: `EFTracking` `IDTPM`
# Performance plots for Pipeline Template
This page provides general instruction to produce performance plots to be included in the Pipeline Template document.
All the histograms can be produced using [IDTPM](https://gitlab.cern.ch/atlas/athena/-/tree/main/InnerDetector/InDetValidation/InDetTrackPerfMon). For detailed information on the IDTPM package and how to run it, see [InDetTrackPerfMon (IDTPM) Package](https://codimd.web.cern.ch/XklVEplsSo-aNroVTrvIiw).
All the performance studies should be performed both with **FullScan** and with **Trigger navigation**. The following instructions apply to FullScan, but trigger navigation can be included by adding the ```ChainNames``` flag to the TrackAnalysis config file (see the [documentation](https://codimd.web.cern.ch/XklVEplsSo-aNroVTrvIiw#Individual-TrackAnalysis-flags))
### TrackAnalysis
To assess the pipeline performance, two "categories" of TrackAnalyses should be run: one for offline (Athena Legacy, C-000) monitoring (```TrkAnaOffl```) and one for EF pipeline (X-XXX) monitoring (```TrkAnaEF```). In addition, a third category can be used for direct comparison between C-000 and X-XXX tracks ((```DoubleRatioTrkAna```)):
1. **TrkAnaOffl**:
* test = C-000 (Athena legacy) tracks container
* reference = truth particles container
* Test-reference matching with MatchingType = TruthMatch
3. **TrkAnaEF**:
* test = X-XXX tracks container
* reference = truth particles container
* Test-reference matching with MatchingType = TruthMatch
* Need to specify the EF tracks container through the ```TrigTrkKey``` flag
4. **DoubleRatioTrkAna**:
* test = X-XXX tracks container
* reference = offline (Athena legacy) tracks container
* Test-reference matching with MatchingType = EFTruthMatch: the test is considered matched to reference if both are linked to the same truth (with the "TruthMatch" criterion)
* Need to specify the EF tracks container through the ```TrigTrkKey``` flag
A root file will be created, with one subdirectory for each TrackAnalysis. Each of these subdirectories will have the following structure:
* TrackAnalysisName
* Offline (for FullScan mode, trigger chain for trigger navigation)
* Tracks
* Parameters
* ...
* Efficiencies
* ...
* FakeRates
* ...
* Multiplicities
* ...
The ```Offline/Tracks/Multiplicities/summary``` histogram (shown in figure) summarized the number of events, and the number of all/selected/matched test and reference tracks.
### Configuration for Full-scan TrackAnalysis
Example of TrackAnalysis config file:
```
{
"TrkAnaEF" : {
"enabled" : true,
"TestType" : "EFTrigger",
"RefType" : "Truth",
"TrigTrkKey": "xAODPipelineTrackCollection", #container of pipeline tracks in xAOD TrackParticle format
"MatchingType" : "TruthMatch",
"plotDuplicateRates": true,
"plotTechnicalEfficiencies" : true
},
"TrkAnaOffl" : {
"enabled" : true,
"TestType" : "Offline",
"RefType" : "Truth",
"MatchingType" : "TruthMatch"
"plotDuplicateRates": true,
"plotTechnicalEfficiencies" : true
},
"TrkAnaDoubleRatio" : {
"enabled" : true,
"TestType" : "EFTrigger",
"RefType" : "Offline",
"TrigTrkKey": "xAODPipelineTrackCollection",
"MatchingType" : "EFTruthMatch",
"SelectOfflineObject" : "Truth",
"plotTrackParameters" : false
}
}
```
### Configuration for Trigger Navigation TrackAnalysis
### Truth object selection (PRELIMINARY)
In addition to the "baseline" configuration, for each TrackAnalysis a set of selection on the truth particles or reconstructed tracks can be applied by setting the corresponding flags. The list of all available flags can be found [here](https://gitlab.cern.ch/atlas/athena/-/blob/main/InnerDetector/InDetValidation/InDetTrackPerfMon/python/InDetTrackPerfMonFlags.py#L57-L125). In particular, the ```"SelectTruthObject"``` flag can be used to apply a default set of truth selections depending on the sample and truth particle considered:
1. ```"SelectTruthObject"="highPTMuon"```: pdgid = 13, pT > 10 GeV
2. ```"SelectTruthObject"="lowPTMuon"```: pdgid = 13, 1 GeV < pT < 10 GeV
3. ```"SelectTruthObject"="highPTPion"```: isHadron = True, pT > 10 GeV
4. ```"SelectTruthObject"="lowPTPion"```: isHadron = True, 1 GeV < pT < 10 GeV
5. ```"SelectTruthObject"="highPTElectron"```: pdgid = 11, pT > 20 GeV
6. ```"SelectTruthObject"="lowPTElectron"```: pdgid = 11, 10 GeV < pT < 20 GeV
***To do: add list of flags for offline track selections, including WPs. Add corresponding column in the plot tables***
### Plotting
Once the IDTPM histograms are created, the post-processing can be done using [DCube](https://gitlab.cern.ch/art/dcube/-/blob/master/README.md), which can easily compare all histograms from two root files, if the two files have the same structure. An intermediate step will therefore be needed, to convert the IDTPM root file containing results of all the TrackAnalyses into two separate files: one containing the TrkAnaEF and the TrkAnaDoubleRatio (dcube "monitored" file), the other one containing the TrkAnaOffl (renamed as the TrkAnaEF) (dcube reference file). This way, Dcube will produce plots with comparison between TrkAnaOffl and TrkAnaEF, and plots for TrkAnaDoubleRatio with no reference.
An example run command:
```
dcube.py
-p -x output_dir -c config.xml \
--plotopts=ratio \
-r reference.root \
-R 'ref = C-000 -M 'mon = X-XXX’ \
monitored.root
```
The config.xml is the dcube config file, where the list of needed plots can be configured.
A standalone, more flexible script will also be provided for the "final" plots to include in the template document.
### The global workflow
The global workflow can be summarized in the following 3 steps:
1. Run IDTPM, for all the needed TrackAnalyses => create one root file containing monitoring histograms for C-000, EF, and DoubleRatio;
2. Run an "output-conversion" script that splits the IDTPM root file into the monitored and reference files;
3. Run DCube to produce the plots and save them in a web page for easy visualization;
An ART test should be prepared for each X-XXX, to allow for day-by-day monitoring.
### ART test
Two types of ART tests will be made available depending on whether or not the pipeline workflow allows to run C-000 and X-XXX tracking and store them both in the same output file in separate TrackParticle containers, therefore enabling the usage of ```TrkAnaDoubleRatio```.
The results of all IDTPM ART tests can be found [here](https://https://bigpanda.cern.ch/art/jobs/?package=InDetTrackPerfMon&nlastnightlies=7&view=packages&branch=main/Athena/x86_64-el9-gcc13-opt)
The links to the available ART scripts (which use ttbar PU=200 samples) are reported in the following table:
| Pipeline | Full-scan | Trigger Navigation |
| ----------- |:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------ |
| C-000 (ref) | [ttbar_PU200](https://gitlab.cern.ch/atlas/athena/-/blob/main/InnerDetector/InDetValidation/InDetTrackPerfMon/test/test_EFTrack_C000_FS_ttbar_pu200_grid.sh)| |
| C-100 | [ttbar_PU200](https://gitlab.cern.ch/atlas/athena/-/blob/main/InnerDetector/InDetValidation/InDetTrackPerfMon/test/test_EFTrack_C100_FS_ttbar_pu200_grid.sh) | |
| C-200 | | |
| G-100 | | |
| G-200 | [ttbar_PU200](https://gitlab.cern.ch/atlas/athena/-/blob/main/InnerDetector/InDetValidation/InDetTrackPerfMon/test/test_EFTrack_G200_FS_ttbar_pu200_grid.sh) | |
| G-300 | | |
| G-400 | | |
| F-100 | [ttbar_PU200](https://gitlab.cern.ch/atlas/athena/-/blob/main/InnerDetector/InDetValidation/InDetTrackPerfMon/test/test_EFTrack_F100_FS_ttbar_pu200_grid.sh) | |
| F-400 | | |
| F-600 | | |
# List of relevant numbers and plots
## Efficiencies
Efficiencies in IDTPM are stored as ```TEfficiency``` objects showing the $N_{matched}/N_{selected}$ ratio for truth and reconstructed (X-XXX or C-000) tracks, in bins of a set of variables of interest. One (C-000 or X-XXX) reconstructed track or truth is considered **matched** if least one reconstructed-truth matching exists.
The naming convention is ```eff_vs_[test/ref]_[var]```, where ```[var] ``` indicates the variable of interest. The actual tracking efficiencies are ```eff_vs_[ref]_[var]```, while the ```eff_vs_[test]_[var]``` are effectively purity.
### Requirements and performance assessment
> ***Requirement 2.5*** *The minimum tracking efficiency requirement for high-pT tracks above 10 GeV (low-pT tracking between 1-10 GeV) shall be 98% (95%), as measured with respect to offline tracks. This requirement shall apply to muons and pions.*
> ***Requirement 2.6*** *The minimum tracking efficiency requirement for high-pT tracks above 20 GeV (lower-pT tracking between 10-20 GeV) shall be 98% (97%), as measured with respect to offline tracks. This requirement shall apply to electrons.*
The efficiency performance should be evaluated with respect to C-000 tracking performance. This can be done by comparing the efficiencies for the two TrackAnalyses, ```TrkAnaOffl``` and ```TrkAnaEF```.
The minimum requirements translate into showing that the ratio of X-XXX and C-000 inclusive efficiencies (full-pt and low or high pt regions) is above the given thresholds. Further details related to the performance as a function of the kinematic variables should be provided as summarized in the following tables, showing the efficiency comparison for the two TrackAnalyses and the X-XXX/C-000 ratio.
An alternative strategy to produce X-XXX/C-000 ratios, which could be used as an additional cross-check or an easier way to visualize the performance comparison directly from the IDTPM root file, is using the ```TrkAnaDoubleRatio```. Note that the ```TrkAnaDoubleRatio``` requires also Athena legacy tracks (```ITkInDetTrackParticles```) to be stored in the same AOD file as the X-XXX tracks.
The two methodologies are expected to provide consistent results.
**Technical efficiency** plots are also available, considering only "reconstructable" truth particles, based on the associated number of hits as resulting from C100 clustering (ACTS fast tracking). Technical efficiencies can be estracted replacing ```Efficiencies/eff_vs``` with ```Efficiencies/Technical/tech_eff_vs``` in the IDTPM histogram name.
### Main Infos
Provide inclusive efficiencies (full pT range and high / low pT) in a table (see the template). These can be extracted from the inclusive IDTPM plots listed in the table, and are included in the SummaryTable that is printed out as part of the ART test output.
| pT range | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject | Samples |
| --------- | ----------------------------------- | ----------------------- | ------------------- | --- |
| Inclusive | Efficiencies/eff_vs_truth_inclusive | TrkAnaEF TrkAnaOffl | All | $t\bar{t}$ semi-lep, $Z\to\mu\mu$, $Z\to ee$|
| Low pT | Efficiencies/eff_vs_truth_inclusive | TrkAnaEF TrkAnaOffl | lowPTMuon lowPTElectron | $Z\to\mu\mu$ $Z\to ee$|
| High pT | Efficiencies/eff_vs_truth_inclusive | TrkAnaEF TrkAnaOffl | highPTMuon highPTElectron | $Z\to\mu\mu$ $Z\to ee$ |
Note that the separation in high and low pT region applies only to electrons and muons, while full-pT results alone should be provided for all remaining signatures/particle types.
The following plots should be included in the report.
| Plot (X-XXX vs C-000 comparison + C-000/X-XXX) | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject | Sample |
| ------------------------------------------------------ | ----------------------------- | ------------------------ | ------------------- | --- |
| Efficiency vs truth PT | Efficiencies/eff_vs_truth_pt | TrkAnaEF TrkAnaOffl | All | $t\bar{t}$ semi-lep, $Z\to\mu\mu$, $Z\to ee$ |
| Efficiency vs truth eta | Efficiencies/eff_vs_truth_eta | TrkAnaEF TrkAnaOffl | Charged Pion / All | $t\bar{t}$ semi-lep |
| Efficiency vs truth eta in low pT region | Efficiencies/eff_vs_truth_eta | TrkAnaEF TrkAnaOffl | highPTMuon lowPTElectron | Single $\mu^{\pm}$ ($p_T=1,10$ GeV), $Z\to\mu\mu$ Single $e^{\pm}$ ($p_T=10$ GeV), $Z\to ee$ |
| Efficiency vs truth eta in high pT region | Efficiencies/eff_vs_truth_eta | TrkAnaEF TrkAnaOffl | highPTMuon highPTElectron | Single $\mu^{\pm}$ ($p_T=10, 100$ GeV), $Z\to\mu\mu$ Single $e^{\pm}$ ($p_T=100$ GeV), $Z\to ee$ |
### Additional plots (appendix?)
| Plot (X-XXX vs C-000 comparison + C-000/X-XXX) | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject | Samples |
| --------------------------------------------------------------------- | --------------------------------- | ----------------------- | ------------------- |--- |
| Efficiency vs muon/electron **var** = [phi, d0, z0] in low pT region | Efficiencies/eff_vs_truth_**var** | TrkAnaEF TrkAnaOffl | lowPTMuon lowPTElectron |
| Efficiency vs muon/electron **var** = [phi, d0, z0] in high pT region | Efficiencies/eff_vs_truth_**var** | TrkAnaEF TrkAnaOffl | highPTMuon highPTElectron |
| Efficiency vs truth **var** = [phi, d0, z0] | Efficiencies/eff_vs_truth_**var** | TrkAnaEF TrkAnaOffl | All |
----
## Resolutions
The resolution of X as a function of Y, (```resolution_[X]_vs_[ref]_[Y]```) is defined as the RMS (after removing outliers) of the X residual ($X_{matched}^{[test]}-X_{matched}^{[ref]}$ distribution in bins of the Y variable.
### Requirements
> ***Requirement 2.7*** *The minimum resolution requirement on the track parameter pT shall be within a factor of two with respect to the offline track pT resolution*
> ***Requirement 2.8*** *The minimum resolution requirement on the track impact parameters (d0 and z0) shall be within a factor of two with respect to the offline track impact parameter resolution.*
This requirements can be assessed showing that the ratio of resolution plots produced with ```TrkAnaOffl``` and ```TrkAnaEF``` is consistently below 2.
### Main Infos
Provide inclusive resolutions of pT, d0 and z0 in a table (see the template). These can be extracted from the inclusive IDTPM plots as described in the following table
| Resolution of variable | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject | Samples |
| ---------------------- | -------------------------------------------- | ----------------------- | ----------------- | --- |
| pT | Resolutions/resolution_pt_vs_truth_inclusive | TrkAnaEF TrkAnaOffl | All | $t\bar{t}$ semi-lep, $Z\to\mu\mu$, $Z\to ee$ |
| d0 | Resolutions/resolution_d0_vs_truth_inclusive | TrkAnaEF TrkAnaOffl | All | $t\bar{t}$ semi-lep, $Z\to\mu\mu$, $Z\to ee$|
| z0 | Resolutions/resolution_z0_vs_truth_inclusive | TrkAnaEF TrkAnaOffl | All | $t\bar{t}$ semi-lep, $Z\to\mu\mu$, $Z\to ee$|
The pt, d0 and z0 resolutions should be shown, as a function of pt and eta.
| Plot (X-XXX vs C-000 comparison + C-000/X-XXX) | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject | Samples |
| ------------------------------------------------------- | -------------------------------------- | ------------------------- |----------------- | --- |
| pt resolution vs truth eta | Resolutions/resolution_pt_vs_truth_eta | TrkAnaOffl TrkAnaEF | All | |
| d0 resolution vs truth eta | Resolutions/resolution_d0_vs_truth_eta | TrkAnaOffl TrkAnaEF | All ||
| z0 resolution vs truth eta | Resolutions/resolution_z0_vs_truth_eta | TrkAnaOffl TrkAnaEF | All | |
| pt resolution vs truth pt | Resolutions/resolution_pt_vs_truth_pt | TrkAnaOffl TrkAnaEF | All | |
| d0 resolution vs truth pt | Resolutions/resolution_d0_vs_truth_pt | TrkAnaOffl TrkAnaEF | All | |
| z0 resolution vs truth pt | Resolutions/resolution_z0_vs_truth_pt | TrkAnaOffl TrkAnaEF | All | |
### Additional Plots (appendix?)
| Plot (X-XXX vs C-000 comparison + C-000/X-XXX) | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject |
| -------------------------------------------------------- | --------------------------------------- | ------------------------- | ----------------- |
| eta resolution vs truth eta | Resolutions/resolution_eta_vs_truth_eta | TrkAnaOffl TrkAnaEF | All |
| eta resolution vs truth pt | Resolutions/resolution_eta_vs_truth_pt | TrkAnaOffl TrkAnaEF | All |
| phi resolution vs truth pt | Resolutions/resolution_phi_vs_truth_pt | TrkAnaOffl TrkAnaEF | All |
| phi resolution vs truth eta | Resolutions/resolution_phi_vs_truth_eta | TrkAnaOffl TrkAnaEF | All |
*Residuals and pulls*
## Fake rates and duplicate
Fake and duplicate rates are stored as TEfficiency objects filled respectively with $N^{reco}_{matchProb<0.5}/N^{reco}_{selected}$ (```fakerate_vs_[trkType]_[var]```) and $N^{truth}_{matchProb>0.5}/N^{truth}_{selected}$ (```duplicate_vs_truth_[var]```) as a function of a set of variables of interest. Truth and C-000/pipeline track multiplicities per event are stored as well, in the Multiplicities subdirectory.
### Requirements
> ***Definition 5.7*** *Fake (and duplicate) track rate is defined inclusively as the difference in each RR between the number of reconstructed tracks passing the track selection, as defined in Definition 5.1, and the number of generated primary charged particles passing selection criteria. This difference is then normalized to the total number of reconstructed tracks. Pure fakes are defined as those with a matching probability Pmatch ≤ 0.5, while the remaining tracks are considered as duplicates.*
> ***Requirement 2.9*** *The maximum fake and duplicate tracks rate at μ = 200 shall not exceed a factor of 1% times the number of offline reconstructed tracks. This includes the case of track counting in a ∆R cone around a reference particle.*
>
This requirement can be assessed showing that the relative difference of the mean values of track multiplicity plots produced with ```TrkAnaOffl``` and ```TrkAnaEF``` is below 1%. More detailed plots of actual fake and duplicate rates should be shown.
Results will be summarized in a table reporting, for truth, X-XXX, and C-000 tracks, the selected/matched/un-matched tracks multiplicities, followed by the inclusive fake+duplicate, fake and duplicate rates. The rates can be obtained directly from the multiplicities as detailed in the template table. Multiplicities themselves can be extracted from the multiplicity histograms in the IDTPM root file.
### Main infos
| Value | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject |
| ------------------------- |:-------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------- | ----------------- |
| number of selected tracks | mean of Multiplicities/num_truth_selected Multiplicities/num_eftrig_selected Multiplicities/num_offl_selected | .. TrkAnaOffl TrkAnaOffl TrkAnaEF | All |
| number of matched tracks | mean of Multiplicities/num_truth_matched Multiplicities/num_eftrig_matched Multiplicities/num_offl_matched | .. TrkAnaOffl TrkAnaOffl TrkAnaEF | All |
Show the following plots
| Plot (comparison + ratio ) | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject |
| ---------------------------------------- | ------------------------------------------------------------------------ | ----------------------- | ----------------- |
| Number of selected reco tracks per event | Multiplicities/num_truth_selected Multiplicities/num_eftrig_selected | TrkAnaEF | All |
| Number of matched reco tracks per event | Multiplicities/num_truth_selected Multiplicities/num_offl_selected | TrkAnaOffl | All |
| Fake rate vs offline/pipeline pt | FakeRates/fakerate_vs_offl_pt FakeRates/fake_vs_eftrig_pt | TrkAnaOffl TrkAnaEF | All |
| Duplicate rate vs truth pt | FakeRates/duplicate_vs_truth_pt | TrkAnaOffl TrkAnaEF | All |
### Additional plots (appendix?)
| Plot (X-XXX vs C-000 comparison + C-000/X-XXX) | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject |
| ------------------------------------------------------ | ------------------------------------------------------------------- | ----------------------- | ----------------- |
| Fake rate vs offline/pipeline eta | FakeRates/fakerate_vs_offl_eta FakeRates/fakerate_vs_eftrig_eta | TrkAnaOffl TrkAnaEF | All |
| Duplicate rate vs truth eta | FakeRates/duplicate_vs_truth_eta | TrkAnaOffl TrkAnaEF | All |
# Robustness versus PU (and defects)
The tracking performance robustness versus PU condition should be proven comparing three scenarios with increasing PU, $\mu=0$, $\mu=140$ and $\mu=200$.The efficiency, pt, d0 and z0 resolution, fake and duplicate rates versus truth pT should be shown. Similarly for robustness against defects, using samples that will be made available by PPES.
| Plot (PU0 vs PU140 vs PU200 comparison + C-000/X-XXX) | IDTPM histo | IDTPM TrackAnalysis | SelectTruthObject |
| -------------------------------------------------------------- | ------------------------------------------------------------- | ------------------- | ----------------- |
| pt resolution vs truth pt | Resolutions/resolution_pt_vs_truth_pt | TrkAnaEF | All |
| d0 resolution vs truth pt | Resolutions/resolution_d0_vs_truth_pt | TrkAnaEF | All |
| z0 resolution vs truth pt | Resolutions/resolution_z0_vs_truth_pt | TrkAnaEF | All |
| Efficiency vs truth PT | Efficiencies/eff_vs_truth_pt | TrkAnaEF | All |
| Fake rate vs offline/pipeline pt | FakeRates/fakerate_vs_offl_pt FakeRates/fake_vs_eftrig_pt | TrkAnaEF | All |
| Duplicate rate vs truth pt | FakeRates/duplicate_vs_truth_pt | TrkAnaEF | All |
# Track parameters (appendix?)
... to do ...
# Samples
Samples production for Trigger Upgrade studies: [ATLMCPROD-11441](https://its.cern.ch/jira/browse/ATLMCPROD-11441)
* Geometry: ATLAS-P2-RUN4-03-00-01
* e8557
* s4345 (Athena,25.0.03 with conditions OFLCOND-MC21-SDR-RUN4-02)
* r16128, 16129, 16130 (PU = 0, 140, 200)
| Process | DSID | Sample | Particle type |
| ----------------------------------------- | ------------ |:----------------------------------------- |:------------------------------------- |
| Single $e^{\pm}$ ($p_{T}=10/100$ GeV) | 900494/7 | PG_single_epm_Pt10/100_etaFlatnp0_43 | Electron |
| Single $\mu^{\pm}$ ($p_{T}=1/10/100$ GeV) | 900492/5/8 | PG_single_muonpm_Pt1/10/100_etaFlatnp0_43 | Muon |
| Single $\pi^{\pm}$ ($p_{T}=1/10/100$ GeV) | 900034/38/42 | PG_singlepiplus_Pt1/10/100_etaFlatnp0_43 | Charged Pion |
| $Z\rightarrow ee$ | 601189 | PhPy8EG_AZNLO_Zee | Electron |
| $Z\rightarrow \mu\mu$ | 601190 | PhPy8EG_AZNLO_Zmumu | Muon |
| $t\bar{t}$ semi-lep. | 601229 | PhPy8EG_A14_ttbar_hdamp258p75_SingleLep | Charged Pion / All (Inclusive sample) |
:::spoiler Additional Samples
| Process | DSID | Sample | Particle type |
| ----------------- | ------ | ------------------------------------ | ------------- |
| $t\bar{t}$ allhad | 601237 | PhPy8EG_A14_ttbar_hdamp258p75_allhad | Jets |
| HH->(bb)(tautau) | 603700 | PhPy8EG_HHbbttHadHad_cHHH01d0 | Tau |
| HH->4b | 603404 | PhPy8EG_HH4b_cHHH01d0 | b-jets |
:::
-----
# Check-list
- [ ] AOD file completeness
- [ ] truth hits info (needed for technical efficiencies):
```--preExec 'flags.Tracking.writeExtendedSi_PRDInfo=True'```
- [ ] truth links (needed for truth matching)
- [ ] Selections
- [ ] No selections applied to truth particles and offline tracks at the pipeline level (all selections should be handled at the IDTPM level)
- [ ] No selection applied to EF tracks at the pipeline level, unless strictly needed (to be discussed with Performance coordinators)
- [ ] Common algorithms consistency: check that any step that is shared with other pipelines is implemented in a perfectly consistent way
- [ ] Tracking in RoI (Trigger Navigation)