Skip to content

SomaDataIO

The SomaDataIO R package loads and exports ‘SomaScan’ data via the SomaLogic Operating Co., Inc. structured text file called an ADAT (*.adat). The package also exports auxiliary functions for manipulating, wrangling, and extracting relevant information from an ADAT object once in memory. Basic familiarity with the R environment is assumed, as is the ability to install contributed packages from the Comprehensive R Archive Network (CRAN).

If you run into any issues/problems with SomaDataIO full documentation of the most recent release can be found at our website of articles and workflows. If the issue persists we encourage you to consult the issues page and, if appropriate, submit an issue and/or feature request.

Usage

The SomaDataIO package is licensed under the MIT license and is intended solely for research use only (“RUO”) purposes. The code contained herein may not be used for diagnostic, clinical, therapeutic, or other commercial purposes.

Installation

The easiest way to install SomaDataIO is to install directly from CRAN:

install.packages("SomaDataIO")

Alternatively from GitHub:

remotes::install_github("SomaLogic/SomaDataIO")

which installs the most current “development” version from the repository HEAD. To install the most recent release, use:

remotes::install_github("SomaLogic/SomaDataIO@*release")

To install a specific tagged release, use:

remotes::install_github("SomaLogic/SomaDataIO@v5.3.0")

Package Dependencies

The SomaDataIO package was intentionally developed to contain a limited number of dependencies from CRAN. This makes the package more stable to external software design changes but also limits its contained feature set. With this in mind, SomaDataIO aims to strike a balance providing long(er)-term stability and a limited set of features. Below are the package dependencies (see also the DESCRIPTION file):

Biobase

The Biobase package is suggested, being required by only two functions, pivotExpressionSet() and adat2eSet(). Biobase must be installed separately from Bioconductor by entering the following from the R Console:

if (!requireNamespace("BiocManager", quietly = TRUE)) {
  install.packages("BiocManager")
}
BiocManager::install("Biobase", version = remotes::bioc_version())

Information about Bioconductor can be found here: https://bioconductor.org/install/

Loading

Upon successful installation, load the SomaDataIO as normal:

For an index of available commands:

library(help = SomaDataIO)

Objects and Data

The SomaDataIO package comes with four (4) objects available to users to run canned examples (or analyses). They can be accessed once SomaDataIO has been attached via library(). They are:

  • example_data: the original ‘SomaScan’ file (example_data.adat) can be found here or downloaded directly via:

    wget https://raw.githubusercontent.com/SomaLogic/SomaLogic-Data/main/example_data.adat

    • within SomaDataIO it has been replaced by an abbreviated, light-weight version containing only the first 10 samples:

      dir(system.file("extdata", package = "SomaDataIO"), full.names = TRUE)

  • ex_analytes: the analyte (feature) variables in example_data

  • ex_anno_tbl: the annotations table associated with example_data

  • ex_target_names: a mapping object for analyte -> target

  • See also ?SomaScanObjects

Main (I/O) Features

  • Loading data (Import)
    • parse and import a *.adat text file into an R session as a soma_adat object.
  • Wrangling data (manipulation)
  • Exporting data (Output)
    • write out a soma_adat object as a *.adat text file.

Loading an ADAT

Loading an ADAT text file is simple using read_adat():

# Sample file name
f <- system.file("extdata", "example_data10.adat",
                 package = "SomaDataIO", mustWork = TRUE)
my_adat <- read_adat(f)

# test object class
is.soma_adat(my_adat)
#> [1] TRUE

# S3 print method (forwards -> tibble)
my_adat
#> ══ SomaScan Data ═══════════════════════════════════════════════════════════════
#>      SomaScan version     V4 (5k)
#>      Signal Space         5k
#>      Attributes intact    ✓
#>      Rows                 10
#>      Columns              5318
#>      Clinical Data        34
#>      Features             5284
#> ── Column Meta ─────────────────────────────────────────────────────────────────
#> ℹ SeqId, SeqIdVersion, SomaId, TargetFullName, Target, UniProt, EntrezGeneID,
#> ℹ EntrezGeneSymbol, Organism, Units, Type, Dilution, PlateScale_Reference,
#> ℹ CalReference, Cal_Example_Adat_Set001, ColCheck,
#> ℹ CalQcRatio_Example_Adat_Set001_170255, QcReference_170255,
#> ℹ Cal_Example_Adat_Set002, CalQcRatio_Example_Adat_Set002_170255, Dilution2
#> ── Tibble ──────────────────────────────────────────────────────────────────────
#> # A tibble: 10 × 5,319
#>    row_names      PlateId  PlateRunDate ScannerID PlatePosition SlideId Subarray
#>    <chr>          <chr>    <chr>        <chr>     <chr>           <dbl>    <dbl>
#>  1 258495800012_3 Example… 2020-06-18   SG152144… H9            2.58e11        3
#>  2 258495800004_7 Example… 2020-06-18   SG152144… H8            2.58e11        7
#>  3 258495800010_8 Example… 2020-06-18   SG152144… H7            2.58e11        8
#>  4 258495800003_4 Example… 2020-06-18   SG152144… H6            2.58e11        4
#>  5 258495800009_4 Example… 2020-06-18   SG152144… H5            2.58e11        4
#>  6 258495800012_8 Example… 2020-06-18   SG152144… H4            2.58e11        8
#>  7 258495800001_3 Example… 2020-06-18   SG152144… H3            2.58e11        3
#>  8 258495800004_8 Example… 2020-06-18   SG152144… H2            2.58e11        8
#>  9 258495800001_8 Example… 2020-06-18   SG152144… H12           2.58e11        8
#> 10 258495800004_3 Example… 2020-06-18   SG152144… H11           2.58e11        3
#> # ℹ 5,312 more variables: SampleId <chr>, SampleType <chr>,
#> #   PercentDilution <int>, SampleMatrix <chr>, Barcode <lgl>, Barcode2d <chr>,
#> #   SampleName <lgl>, SampleNotes <lgl>, AliquotingNotes <lgl>,
#> #   SampleDescription <chr>, …
#> ════════════════════════════════════════════════════════════════════════════════

Please see vignette vignette("tips-loading-and-wrangling", package = "SomaDataIO") for more details and options.

Wrangling

The soma_adat class comes with numerous class-specific S3 methods to the most popular dplyr and tidyr generics.

# see full complement of `soma_adat` methods
methods(class = "soma_adat")
#>  [1] [              [[             [[<-           [<-            ==            
#>  [6] $              $<-            anti_join      arrange        count         
#> [11] filter         full_join      getAdatVersion getAnalytes    getMeta       
#> [16] group_by       inner_join     is_seqFormat   left_join      Math          
#> [21] median         merge          mutate         Ops            print         
#> [26] rename         right_join     row.names<-    sample_frac    sample_n      
#> [31] semi_join      separate       slice_sample   slice          summary       
#> [36] Summary        transform      ungroup        unite         
#> see '?methods' for accessing help and source code

Please see vignette vignette("tips-loading-and-wrangling", package = "SomaDataIO") for more details about available soma_adat methods.

ADAT structure

The soma_adat object also contains specific structure that are useful to users. Please also see ?colmeta or ?annotations for further details about these fields.

Typical ‘SomaScan’ Analysis

This section now lives in individual package articles. For further detail please see:

MIT LICENSE