Lift an ADAT Between Assay Versions

The SomaScan platform continually improves its technical processes between assay versions. The primary change of interest is content expansion, and other protocol changes may be implemented including: changing reagents, liquid handling equipment, and well volumes.

Table of SomaScan assay versions:

Version	Commercial Name	Size
V4	5k	5284
v4.1	7k	7596
v5.0	11k	11083

However, for a given analyte, these technical upgrades can result in minute measurement signal differences, requiring a calibration (aka "lifting" or "bridging") to bring RFUs into a comparable signal space. This is accomplished by applying an analyte-specific scalar, a linear transformation, to each analyte RFU measurement (column). If you have an annotations file (*.xlsx) and wish to examine the bridging scalars themselves, please see read_annotations().

Lifting between SomaScan versions no longer requires an annotations file containing lifting scalars. We now enable users to pass a bridge parameter, indicating the direction of the bridge. For example, to "lift" between 11k -> 7k, you must be acting on SomaScan data in 11k RFU space and would pass bridge = "11k_to_7k". Likewise, 7k -> 5k requires bridge = "7k_to_5k". Lastly, you may also lift directly from 11k -> 5k (aka "double-bridge") with bridge = "11k_to_5k". See below for all options for the bridge argument.

Usage

lift_adat(
  adat,
  bridge = c("11k_to_7k", "11k_to_5k", "7k_to_11k", "7k_to_5k", "5k_to_11k", "5k_to_7k"),
  anno.tbl = deprecated()
)

is_lifted(adat)

Arguments

adat

A soma_adat object (with intact attributes), typically created using read_adat().

bridge

The direction of the lift (i.e. bridge).

anno.tbl

. Please now use the bridge argument.

Value

lift_adat(): A "lifted" soma_adat object corresponding to the scaling requested in the bridge parameter. RFU values are rounded to 1 decimal place to match standard SomaScan delivery format.

is_lifted(): Logical. Whether the RFU values in a soma_adat

have been lifted from its original signal space to a new signal space.

Details

Matched samples across assay versions are used to calculate bridging scalars. For each analyte, this scalar is computed as the ratio of population medians across assay versions. Please see the lifting vignette vignette("lifting-and-bridging", package = "SomaDataIO") for more details.

Lin's CCC

The Lin's Concordance Correlation Coefficient (CCC) is calculated by computing the correlation between post-lift RFU values and the RFU values generated on the original SomaScan version. This CCC estimate is a measure of how well an analyte can be bridged across SomaScan versions. See vignette("lifting-and-bridging", package = "SomaDataIO"). As with the lifting scalars, if you have an annotations file you may view the analyte-specific CCC values via read_annotations(). Alternatively, getSomaScanLiftCCC() retrieves these values from an internal object for both "serum" and "plasma".

Analyte Setdiff

• Newer versions of SomaScan typically have additional content, i.e. new reagents added to the multi-plex assay that bind to additional proteins. When lifting to a previous SomaScan version, new reagents that do not exist in the "earlier" assay version assay are scaled by 1.0, and thus maintained, unmodified in the returned object. Users may need to drop these columns in order to combine these data with a previous study from an earlier SomaScan version, e.g. with collapseAdats().

• In the inverse scenario, lifting "forward" from a previous, lower-plex version, there will be extra reference values that are unnecessary to perform the lift, and a warning is triggered. The resulting data consists of RFU data in the "new" signal space, but with fewer analytes than would otherwise be expected (e.g. 11k space with only 5284 analytes; see example below).

References

Lin, Lawrence I-Kuei. 1989. A Concordance Correlation Coefficient to Evaluate Reproducibility. Biometrics. 45:255-268.

Examples

# `example_data` is SomaScan (V4, 5k)
adat <- head(example_data, 3L)
dim(adat)
#> [1]    3 5318

getSomaScanVersion(adat)
#> [1] "V4"

getSignalSpace(adat)
#> [1] "V4"

# perform 'lift'
lift_11k <- lift_adat(adat, "5k_to_11k")  # warning
#> Warning: There are extra scaling values (5799) in the reference.
#> They will be ignored.

is_lifted(lift_11k)
#> [1] TRUE

dim(lift_11k)
#> [1]    3 5318

# attributes updated to reflect the 'lift'
attr(lift_11k, "Header")$HEADER$SignalSpace
#> [1] "v5.0"

attr(lift_11k, "Header")$HEADER$ProcessSteps
#> [1] "Raw RFU, Hyb Normalization, medNormInt (SampleId), plateScale, Calibration, anmlQC, qcCheck, anmlSMP, Lifting Bridge (5k -> 11k)"