Import AlphaGenome Results — av_import

Reads an AlphaGenome CSV file and creates an AlphaVarSet object. AlphaGenome ships a stable schema. Column names are preserved verbatim (e.g. variant_id, quantile_score, raw_score, output_type, gene_name, scored_interval).

Usage

av_import_csv(file, sep = ",", verbose = TRUE)

Arguments

file: Path to the AlphaGenome CSV file.
sep: Field separator (default ",").
verbose: Logical. If TRUE (default), emit progress messages.

Value

An AlphaVarSet object with all CSV columns preserved.

Examples

example_file <- system.file(
  "extdata", "bcl11a_rs1427407_example.csv",
  package = "AlphaVaR"
)
obj <- av_import_csv(example_file, verbose = FALSE)
obj
#> === AlphaVarSet Object ===
#> Entries:    2,676 
#> Variants:   2 
#> Modalities: 11 
#> Columns:    25 (incl. 19 extra)
#> Score Mode: quantile_score 
#> Source:     bcl11a_rs1427407_example.csv 

# All 25 AlphaGenome columns survive import. Only low-cardinality metadata
# is cast to factor, which keeps large score tables in memory.
dim(obj$data)
#> [1] 2676   25
levels(obj$data$output_type)
#>  [1] "ATAC"              "CAGE"              "CHIP_HISTONE"     
#>  [4] "CHIP_TF"           "CONTACT_MAPS"      "DNASE"            
#>  [7] "PROCAP"            "RNA_SEQ"           "SPLICE_JUNCTIONS" 
#> [10] "SPLICE_SITES"      "SPLICE_SITE_USAGE"

# quantile_score becomes the active score whenever the CSV carries it.
obj$meta$active_score
#> [1] "quantile_score"
av_set_active_score(obj, "raw")$meta$active_score
#> Switched active score: quantile_score -> raw_score. Statistics cleared.
#> [1] "raw_score"
head(av_get_scores(obj))
#> [1] 0.9812546 0.9780051 0.9828931 0.7844907 0.8069733 0.6133556