Skip to content

Latest commit

 

History

History
73 lines (58 loc) · 2.57 KB

symbol_conversion.md

File metadata and controls

73 lines (58 loc) · 2.57 KB
Raw

Converting NicheNet’s model from human to mouse symbols

Robin Browaeys 2019-07-31

In this vignette, we show how to convert NicheNet’s ligand-target matrix model from human to mouse gene symbols. This is necessary if you want to apply NicheNet to mouse expression data, because the NicheNet prior information is given in human gene symbols (because most data sources at the basis of NicheNet are based on human data). One-to-one orthologs were gathered from NCBI HomoloGene and also from ENSEMBL via biomaRt.

Load required packages

library(nichenetr)
library(tidyverse)

Load NicheNet’s ligand-target model:

ligand_target_matrix = readRDS(url("https://zenodo.org/record/3260758/files/ligand_target_matrix.rds"))
ligand_target_matrix[1:5,1:5] # target genes in rows, ligands in columns
##                 CXCL1        CXCL2        CXCL3        CXCL5         PPBP
## A1BG     3.534343e-04 4.041324e-04 3.729920e-04 3.080640e-04 2.628388e-04
## A1BG-AS1 1.650894e-04 1.509213e-04 1.583594e-04 1.317253e-04 1.231819e-04
## A1CF     5.787175e-04 4.596295e-04 3.895907e-04 3.293275e-04 3.211944e-04
## A2M      6.027058e-04 5.996617e-04 5.164365e-04 4.517236e-04 4.590521e-04
## A2M-AS1  8.898724e-05 8.243341e-05 7.484018e-05 4.912514e-05 5.120439e-05
dim(ligand_target_matrix)
## [1] 25345   688

Convert the ligand-target model from human to mouse symbols.

Because not all human genes have a mouse one-to-one ortholog, these genes will be removed from the mouse model.

colnames(ligand_target_matrix) = ligand_target_matrix %>% colnames() %>% convert_human_to_mouse_symbols() 
rownames(ligand_target_matrix) = ligand_target_matrix %>% rownames() %>% convert_human_to_mouse_symbols() 

ligand_target_matrix = ligand_target_matrix %>% .[!is.na(rownames(ligand_target_matrix)), !is.na(colnames(ligand_target_matrix))]

dim(ligand_target_matrix)
## [1] 17330   644

Show the top 10 targets of TNF (in mouse symbols):

top_targets = extract_top_n_targets("Tnf",10,ligand_target_matrix) %>% names()
top_targets
##  [1] "Hacd4"  "P3h2"   "Sele"   "Vcam1"  "Ubd"    "Ccl19"  "Muc5ac"
##  [8] "Cxcl9"  "Crp"    "Icam1"

If you want to convert mouse to human symbols, you can use:

top_targets %>% convert_mouse_to_human_symbols()
##    Hacd4     P3h2     Sele    Vcam1      Ubd    Ccl19   Muc5ac    Cxcl9 
##  "HACD4"   "P3H2"   "SELE"  "VCAM1"    "UBD"  "CCL19" "MUC5AC"  "CXCL9" 
##      Crp    Icam1 
##    "CRP"  "ICAM1"