Cell type evolution and diversity in placozoans

This repository contains all necessary scripts to reproduce the single-cell atlases of the placozoans Trichoplax adhaerens (H1), Trichoplax sp. (H2), Hoilungia hongkongensis (H13), and Cladtertia collaboinventa (H23), presented in our manuscript:

Stepwise emergence of the neuronal gene expression program in early animal evolution, (Najle, Grau-Bové et al., Cell 2023).

Table of contents:

• Analyses
• Browse the data
• Access the data

Analyses

Check instructions in each of these folders.

1. Single-cell atlases for each species, in the results_scatlas/ folder. This covers all steps from mapping to the final metacell clustering and cell type classification for each species, as well as creating expression maps, 2D projections, and dedicated expression plots for certain markers (TFs, GPCRs, etc.).
2. Cross-species comparisons within placozoa and with other metazoan species, in results_crosssps/.
3. Gene module analyses, in results_gene_modules/. This includes the analysis of placozoan-specific gene modules, and the analyses of neural-related gene modules across metazoans.
4. Phylogenomics in results_phylogenomics/. Includes instructions to reproduce our single-gene marker phylogenomics datasets, and run the species trees with IQ-TREE and Phylobayes.

Browse the data

You can browse the results from our analyses in this ShinyApp.

Access the data

We provide the following raw data for each species:

• UMI matrices, which are the files starting with the mat prefix. For all analyses, we used matrices indicated with the it2 suffix (other files correspond to previous iterations, used in intermediate analyses). These are sparse matrices stored as R objects, specifically metacell package-type mat objects. See below for an example of how to load these matrices into R.
• Cell-metacell assignment files indicated with the mc prefix. For all analyses, we used matrices indicated with the it4 suffix (other files correspond to intermediate analyses).
• Two-dimensional porjection objects indicated in the mc2d prefix.
• Other objects used by metacell such as gene statistics, gene sets, cell coclustering data... Please refer to the appropriate scripts in this readme for details.

Each species is referred to with a short acronym, all through this project:

• Tadh for Trichoplax adhaerens H1
• TrH2 for Trichoplax sp. H2
• Hhon for Hoilungia hongkongensis H13
• HoiH23 for Cladtertia collaboinventa H23 (we came up with the label before it was recently renamed from a species in the Hoilungia genus to a genus of its own, apologies for the confusion).

We also provide metacell-cell type assignment tables, with our cell type annotations for each metacell. The final annotations are stored in this folder: results_scatlas/results_metacell_it4/, e.g. this file.

The metacell package contains functions to load metacell-formatted data files, such as cell clusterings in mc objects (which specify which cells in the UMI matrix belong to each metacell cluster).

How to load metacell objects in R:

# from the `results_scatlas/` folder in this repository
# load metacell library
library("metacell")
# initialise metacell database using its relative path:
metacell::scdb_init("data/scdb/", force_reinit = TRUE)

# read the mat object using its ID to refer to it (in this case, the `scdr_Tadh_it2` bit); there ara analogous files for metacells, etc
mat = metacell::scdb_mat("scdr_Tadh_it2")
# sparse matrix available in the mat@mat slot, in this case it contains 16386 genes x 13236 cells
dim(mat@mat)

# likewise, you may want to load a mc object containing cell-to-metacell assignments
mc = metacell::scdb_mc("scdr_Tadh_it4")
# the mc@mc slot is a vector with all cells and their associated metacell
length(mc@mc)
# notice that this vector contains only 13151 cells: not all cells in the mat@mat matrix are classified into a metacell

# if you want to know which cell type is annotated to each metacell, check the mc annotation file
# make sure that you load the same version of the metacell clustering as in the mc object, in this case, it4
ctt = read.table("results_metacell_it4/annotation_mc.Tadh.it4.reordered.tsv", header = TRUE)
head(ctt)
# metacell	cell_type	color	broad_cell_type	metacell_it2
# 1	lipophil	khaki2	lipophil	1
# 2	lipophil	khaki2	lipophil	2
# 3	lipophil	khaki2	lipophil	3
# 4	lipophil	khaki2	lipophil	4
# 5	lipophil	khaki2	lipophil	5

If you have any queries, feel free to let me know in the Issues section.

Enjoy!

    _,,.- ~~--~~~~-.._,
  /                     `---.
  \                         .'
   `~- ,_ ,. ,_, ,., ,., -··
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^