Creating a TI method: Wrapping trajectories

Wouter Saelens

Robrecht Cannoodt

2019-10-10

Common trajectory model

dynwrap always represents trajectories in the same way, as illustrated here with a tree trajectory

## # A tibble: 5 x 4
##   from        to          length directed
##   <chr>       <chr>        <dbl> <lgl>   
## 1 Milestone_A Milestone_B    1   FALSE   
## 2 Milestone_B Milestone_C    2   FALSE   
## 3 Milestone_B Milestone_D    1   FALSE   
## 4 Milestone_C Milestone_E    1   FALSE   
## 5 Milestone_C Milestone_F    1.5 FALSE
## # A tibble: 10 x 3
##    cell_id milestone_id percentage
##    <chr>   <chr>             <dbl>
##  1 Cell_a  Milestone_B      0.0326
##  2 Cell_a  Milestone_D      0.967 
##  3 Cell_b  Milestone_B      0.852 
##  4 Cell_b  Milestone_C      0.148 
##  5 Cell_c  Milestone_B      0.539 
##  6 Cell_c  Milestone_C      0.461 
##  7 Cell_d  Milestone_C      0.708 
##  8 Cell_d  Milestone_F      0.292 
##  9 Cell_e  Milestone_C      0.364 
## 10 Cell_e  Milestone_F      0.636
## # A tibble: 10 x 4
##    cell_id from        to          percentage
##    <chr>   <chr>       <chr>            <dbl>
##  1 Cell_a  Milestone_B Milestone_D     0.967 
##  2 Cell_b  Milestone_B Milestone_C     0.148 
##  3 Cell_c  Milestone_B Milestone_C     0.461 
##  4 Cell_d  Milestone_C Milestone_F     0.292 
##  5 Cell_e  Milestone_C Milestone_F     0.636 
##  6 Cell_f  Milestone_B Milestone_C     0.0563
##  7 Cell_g  Milestone_B Milestone_D     0.887 
##  8 Cell_h  Milestone_B Milestone_C     0.150 
##  9 Cell_i  Milestone_B Milestone_C     0.515 
## 10 Cell_j  Milestone_A Milestone_B     0.406
## # A tibble: 3 x 3
##   divergence_id milestone_id is_start
##   <chr>         <chr>        <lgl>   
## 1 Divergence_1  Milestone_B  TRUE    
## 2 Divergence_1  Milestone_C  FALSE   
## 3 Divergence_1  Milestone_D  FALSE

Direct wrapping

These three objects (with either milestone percentages or progressions) are enough to form a trajectory using add_trajectory.

trajectory <- wrap_data(cell_ids = cell_ids) %>% 
  add_trajectory(
    milestone_network = milestone_network, 
    milestone_percentages = milestone_percentages,
    divergence_regions = divergence_regions
  )

Indirect wrapping

Often, you don’t want to directly output the milestone network and percentages, but want to output an alternative representation that is converted by dynwrap to the common representation:

Check out the reference documentation for an overview and examples of the different wrappers