trackSeq

trackSeq is an analysis regime that integrates live cell time lapse microscopy and real time single cell tracking and quantification with high dimensional molecular endpoint (snapshot) analyses. It provides a gene-wise paired t-testing regime between cell pairs, with known cellular and molecular dynamics and fate differences (and ideally with known kinship) to find systematic expression biases toward this fate.

Data and code below from:
Wehling, Loeffler, Zhang, Kull, Donato, Szczerba, Camargo Ortega, Lee, Moor, Goettgens, Aceto and Schroeder
external page Combining single-cell tracking and omics improves blood stem cell fate regulator identification
Blood 2022, 10.1182/blood.2022016880

external page Link to data and code

Wehling-ACDSeq.zip (2.8 GB). Unpacked folders and sizes:

├───Readme.txt [9 KB]
├───LICENSE.txt [9 KB]
├───01_QC_CODE [48 KB]
├───02_STRATIFICATIONS_CODE [7 MB, contains gene annotation file]
├───03_FIGURES_CODE [128 KB]
├───HelperCode [32 KB]
├───NestorowaDataAndHSCGranddaughters [228 MB]
├───mRNA_IO [1.9 GB]
│   ├───Figures [empty]
│   ├───VennDiagramsAndIntersects [24 MB]
│   ├───Output [527 MB] 
│   ├───Input [1.3 GB]
│   │   ├───cDNAdata [26 KB]
│   │   ├───CellCycle_PCNAvsRNAseq[181 KB]
│   │   ├───Countdata [162 MB]
│   │   ├───Moviedata [275 MB]
│   │   ├───Nestorowa2016-Data [812 MB]
│   │   ├───Pickingdata [35 MB]
│   │   ├───PseudoTime [352 KB]
│   │   ├───StringDBAnalysis [68 MB]
├───Other Data [712 MB]
│   ├───AdhesionTests [63 MB]
│   ├───EndpointStains [97 MB]
│   ├───Itgb4PLA [15 MB]
│   ├───PCNAVENUS [7 MB]
│   ├───SmallMolInhibitors_ACDSeq [554 MB]

trackSeq analyzes cell pairs with known history and kinship.
In our workflow, it is used together with the alerT software to first identify and isolate these cells.

Download and unpack Wehling-ACDSeq.zip above. 

The included R code expects the file and data structure as found within the zip file. The workflow consists of three steps that are executed separately from each other as specified below (Step 1 requires R version 3.6, Steps 2&3 require R version ≥ 4.1).

Step 1: 01_QC_CODE: Integrating tracking with scRNA-Seq data, filtering, and normalization

Execute 010_Analsis.R within the folder 01_QC_CODE as the working directory. 

Mapping of tracking to scRNA-Seq data is commented out and already precomputed (The entire section has been pre-computed). If you want to execute it anyway, please use R 3.6.2 and make sure to use Bioconductor version 3.10 to install scater_1.14.0 and scran_1.14.6. (We noticed a changed output of the function scater::quickPerCellQC() in later versions.) 

For more details on the set up, see requirements.txt. 

Step 2: 02_STRATIFICATIONS_CODE: Perform stratifications and candidate identifications

Execute 021_MatchAndRank.R within 02_STRATIFICATIONS_CODE to perform
·      trackSeq (023_Ranking_trackSeqStrat.R),
·      NO- and TSDStrat (024_Ranking_TechnicalVSBiology_PlusMinusTimeSinceDiv.R)
·      KINStrat (025_Ranking_KinshipStrat.R)

Use R 4.1 and the libraries specified in requirements.txt.

Step 3: 03_FIGURES_CODE: Perform network analysis and plotting

Within 03_FIGURES_CODE first execute 040_STRING_AnalysisANDFigures.R to perform the STRING network analysis and then go through 050_Figures.R to create most figures used in the study. Network analyses herein depend on CYTOSCAPE derived .CSV outputs under mRNA_IO > Input based on the .CYS files under mRNA_IO > Input > StringDBAnalysis.

Use R 4.1 as specified in requirements.txt.

If you use trackSeq, please cite the following paper:
Wehling, Loeffler, Zhang, Kull, Donato, Szczerba, Camargo Ortega, Lee, Moor, Goettgens, Aceto and Schroeder
external page Combining single-cell tracking and omics improves blood stem cell fate regulator identification
Blood 2022, 10.1182/blood.2022016880