The method, termed as targeting coalescent analysis (TCA), computes for all cells of a tissue the average coalescent rate at the monophyletic clades of the target tissue, the inverse of which then measures the progenitor number of the tissue. Any predefined population could be investigated with TCA, independent of pre-set markers.
* I recommend using TarCA.beta which could be of higher computational efficiency.
- Dependent packages: dplyr, tidyr, ape, castor, ggplot2, ggtree, phytools, stringr
- Require R (>= 3.5.0).
install.packages('devtools')
devtools::install_github('shadowdeng1994/TarCA')
Installation would finish in about one minute.
library("TarCA")
- The following files are needed for TarCA.
- A tree file of class "phylo" with node labels.
((Cell_1,((Cell_2,Cell_3)Node_4,(Cell_4,Cell_5)Node_5)Node_3)Node_2,(((Cell_6,Cell_7)Node_8,(Cell_8,Cell_9)Node_9)Node_7,Cell_10)Node_6)Node_1;
- A dataframe with columns TipLabel and TipAnn, representing tip labels on the tree file and corresponding cell annotations.
TipLabel TipAnn Cell_1 O1 Cell_2 O1 Cell_3 O1 Cell_4 O2 Cell_5 O2 Cell_6 O2 Cell_7 O3 Cell_8 O3 Cell_9 O3 Cell_10 O3
- (optional) A path to output the result (e.g. NpResult.RData).
- Effective number of progenitor can be inferred with
Np_Estimator. - Modified algorithm for detection of lineage specific expression upregulation (LEU) can be called with
LEU_Estimator.
- Load exemplar dataset.
load(system.file("Exemplar","Exemplar_TCA.RData",package = "TarCA"))
tmp.tree <- ExemplarData_1$Tree
tmp.ann <- ExemplarData_1$Ann
- Inferring Np with
Np_Estimator.
tmp.result <- Np_Estimator(
Tree = tmp.tree,
Ann = tmp.ann,
Fileout = NULL,
ReturnNp = TRUE
)
**** 1. Check input data.
**** 2. Get treedata file.
**** 3. Get node2tip file.
**** 4. Get pureNode file.
**** 5. Get pureNode2organ file.
**** 6. Get CladeSizeDetail file.
**** 7. Get Np file.
- Then the Np estimation are stoarged in
tmp.result[["EffN"]].
TipAnn CladeSize Total EffN O0 1 (1), 2 (2) 5 5 O1 1 (6), 2 (11), 3 (1), 5 (1) 36 26.2 O2 1 (35), 2 (17), 3 (4), 4 (2), 8 (1) 97 67.5 O3 1 (66), 2 (38), 3 (11), 4 (4), 5 (2), 7 (1) 208 158 O4 1 (50), 2 (24), 3 (6), 4 (3), 5 (1) 133 125 O5 1 (71), 2 (38), 3 (13), 4 (5) 206 197 O6 1 (32), 2 (23), 3 (9), 7 (1) 112 87.5 O7 1 (50), 2 (37), 3 (10), 4 (3), 6 (1) 172 147 O8 1 (5), 2 (3) 11 18.3 O9 1 (12), 2 (1), 3 (2) 20 27.1
This process is estimated to be completed in about 30 seconds.
- Load exemplar dataset.
load(system.file("Exemplar","Exemplar_LEU.RData",package = "TarCA"))
tmp.tree <- ExemplarData_2$Tree
tmp.ann <- ExemplarData_2$Ann
- Inferring Np with
LEU_Estimator.
tmp.result <- LEU_Estimator(
Tree = tmp.tree,
Ann = tmp.ann,
Fileout = NULL,
ReturnNp = TRUE
)
**** 1. Check input data.
**** 2. Get BiasNode file.
**** 3. Filter BiasNode.
**** 4. Plot BiasNode.
**** 5. Get pureNode file.
**** 6. Get pureNode2organ file.
**** 7. Get CladeSizeDetail file.
**** 8. Get Np file.
- Then the Np estimation for subpopulation with expression upregulation are stoarged in
tmp.result[["EffN"]].
TipAnn CladeSize Total EffN FALSE 1 (482) 482 Inf TRUE 1 (23), 2 (5), 4 (2) 41 48.2
- Additionally, you can visualize the LEU on the phylogeny with
tmp.result[["BiasFig"]].
This process is estimated to be completed in about 30 seconds.
Shanjun Deng, [email protected].
When using TarCA please cite:
- Deng S, Gong H, Zhang D, et al. A statistical method for quantifying progenitor cells reveals incipient cell fate commitments[J]. Nature Methods, 2024: 1-12.