R tools for ASV occurrence data in SBDI.

• March 2025: Switched counts table to sparse matrices to handle large datasets from the Insect Biome Atlas project, as previous implementation using data.table caused RAM exhaustion. Sparse matrices are currently applied only to counts, but further use for other objects (asvs, events, emof) may follow.

The asvoccur R package, currently under development, provides tools for unpacking and processing ASV occurrence data and metadata downloaded from the Swedish ASV portal. It enables users to convert condensed DwC archives into ASV table format for easier downstream analysis in R, by using functions that load, merge, and aggregate ASV counts across taxonomic ranks.

install.packages('remotes')
remotes::install_github("biodiversitydata-se/asvoccur")
# or:
# remotes::install_github("biodiversitydata-se/asvoccur@develop")
library(asvoccur)

GitHub installations (local source builds) may occasionally fail. If so:

remove.packages("asvoccur")
# restart R / RStudio
remotes::install_github("biodiversitydata-se/asvoccur", force = TRUE)

data_path <- '/path-to/dataset-folder'
loaded <- load_data(data_path)
merged <- merge_data(loaded)
summed <- sum_by_clade(merged$counts, merged$asvs)
summed_df <- convert_to_df(summed)

?asvoccur::load_data
?asvoccur::merge_data
?asvoccur::sum_by_clade
?asvoccur::convert_to_df