return_feature: Return aSVG Files Relevant to Target Features

In jianhaizhang/spatialHeatmap: spatialHeatmap: Visualizing Spatial Assays in Anatomical Images and Large-Scale Data Extensions

Return aSVG Files Relevant to Target Features

Description

This function parses a collection of aSVG files and returns those containing target features in a data frame. Successful spatial heatmap plotting requires the aSVG features of interest have matching samples (cells, tissues, etc) in the data. To meet this requirement, the returned features could be used to replace target sample counterparts in the data. Alternatively, the target samples in the data could be used to replace matching features in the aSVG through function update_feature. Refer to function spatial_hm for more details on aSVG files.

Usage

return_feature(
  feature,
  species,
  keywords.any = TRUE,
  remote = NULL,
  dir = NULL,
  svg.path = NULL,
  desc = FALSE,
  match.only = TRUE,
  return.all = FALSE
)

Arguments

feature	A vector of target feature keywords (case insentitive), which is used to select aSVG files from a collection. E.g. c('heart', 'brain'). If NA or NULL, all features of all SVG files matching species are returned.
species	A vector of target species keywords (case insentitive), which is used to select aSVG files from a collection. E.g. c('gallus'). If NA or NULL, all SVG files in dir are queried.
keywords.any	Logical, TRUE or FALSE. Default is TRUE. The internal searching is case-insensitive. The space, dot, hypen, semicolon, comma, forward slash are treated as separators between words and not counted in searching. If TRUE, every returned hit contains at least one word in the feature vector and at least one word in the species vector, which means all the possible hits are returned. E.g. "prefrontal cortex" in "homo_sapiens.brain.svg" would be returned if feature=c('frontal') and species=c('homo'). If FALSE, every returned hit contains at least one exact element in the feature vector and all exact elements in the species vector. E.g. "frontal cortex" rather than "prefrontal cortex" in "homo_sapiens.brain.svg" would be returned if feature=c('frontal cortex') and species=c('homo sapiens', 'brain').
remote	Logical, FALSE or TRUE. If TRUE (default), the remote EBI aSVG repository https://github.com/ebi-gene-expression-group/anatomogram/tree/master/src/svg and spatialHeatmap aSVG Repository https://github.com/jianhaizhang/SHM_SVG/tree/master/SHM_SVG_repo developed in this project are queried.
dir	The directory path of aSVG files. If remote is TRUE, the returned aSVG files are saved in this directory. Note existing aSVG files with same names as returned ones are overwritten. If remote is FALSE, user-provided (local) aSVG files should be saved in this directory for query. Default is NULL.
svg.path	The path of a specific aSVG file. If the provided aSVG file exists, only features of this file are returned and there will be no querying process. Default is NULL.
desc	Logical, FALSE or TRUE. Default is FALSE. If TRUE, the feature descriptions from the R package "rols" (Laurent Gatto 2019) are added. If too many features are returned, this process takes a long time.
match.only	Logical, TRUE or FALSE. If TRUE (default), only target features are returned. If FALSE, all features in the matching aSVG files are returned, and the matching features are moved on the top of the data frame.
return.all	Logical, FALSE or TRUE. Default is FALSE. If TRUE, all features together with all respective aSVG files are returned, regardless of feature and species.

Value

A data frame containing information on target features and aSVGs.

Author(s)

Jianhai Zhang [e-mail]
Dr. Thomas Girke [e-mail]

References

Laurent Gatto (2019). rols: An R interface to the Ontology Lookup Service. R package version 2.14.0. http://lgatto.github.com/rols/
Hadley Wickham, Jim Hester and Jeroen Ooms (2019). xml2: Parse XML. R package version 1.2.2. https://CRAN.R-project.org/package=xml2
R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
Cardoso-Moreira, Margarida, Jean Halbert, Delphine Valloton, Britta Velten, Chunyan Chen, Yi Shao, Angélica Liechti, et al. 2019. "Gene Expression Across Mammalian Organ Development." Nature 571 (7766): 505-9

Examples

# This function is able to work on the EBI aSVG repository directly: https://github.com/
# ebi-gene-expression-group/anatomogram/tree/master/src/svg. The following shows how to
# download a chicken aSVG containing spatial features of 'brain' and 'heart'. An empty
# directory is recommended so as to avoid overwriting existing SVG files. 
# Here "~/test" is used. 


# Make an empty directory "~/test" if not exist.
if (!dir.exists('~/test')) dir.create('~/test')
# Remote aSVG repos.
data(aSVG.remote.repo)
tmp.dir <- normalizePath(tempdir(check=TRUE), winslash="/", mustWork=FALSE)
tmp.dir.ebi <- paste0(tmp.dir, '/ebi.zip')
tmp.dir.shm <- paste0(tmp.dir, '/shm.zip')
# Download the remote aSVG repos as zip files. According to Bioconductor's 
# requirements, downloadings are not allowed inside functions, so the repos are 
# downloaded before calling "return_feature".  
download.file(aSVG.remote.repo$ebi, tmp.dir.ebi)
download.file(aSVG.remote.repo$shm, tmp.dir.shm)
remote <- list(tmp.dir.ebi, tmp.dir.shm)
# Query the remote aSVG repos.
feature.df <- return_feature(feature=c('heart', 'brain'), species=c('gallus'), dir='~/test',
match.only=FALSE, remote=remote)
feature.df
# The path of downloaded aSVG.
svg.chk <- '~/test/gallus_gallus.svg'


# The spatialHeatmap package has a small aSVG collection and can be used to demonstrate the
# local query.
# Get the path of local aSVGs from the package.
svg.dir <- system.file("extdata/shinyApp/data", package="spatialHeatmap")
# Query the local aSVG repo. The "species" argument is set NULL on purpose so as to illustrate
# how to select the target aSVG among all matching aSVGs.
feature.df <- return_feature(feature=c('heart', 'brain'), species=NULL, dir=svg.dir, 
match.only=FALSE, remote=NULL)
# All matching aSVGs.
unique(feature.df$SVG)
# Select the target aSVG of chicken.
subset(feature.df, SVG=='gallus_gallus.svg')

jianhaizhang/spatialHeatmap documentation built on Nov. 28, 2024, 4:44 p.m.