Score the CoDaCoRe stagewise log-ratio-balance discriminator (pure base R)

Scores each row of X independently with the frozen model from fit_coda_codacore, in PURE base R (no python). Each query is aligned to the frozen feature universe; for each frozen discrete balance $b$ the specimen's discrete ILR balance is computed over its OWN strictly-positive support in $A_b / B_b$ (a side with no positive feature contributes the neutral value 0); the score is the frozen logistic linear predictor $\mathrm{intercept} + \sum_b w_b\,\mathrm{bal}_b(x)$. Larger = more case-like.

Queries with fewer than model$hp$min_features universe features present (a column-overlap floor), or empty positive support over the frozen universe on the aligned ORIGINAL abundances (!any(X_use[i, ] > 0)), return the neutral score 0. A FLAT all-equal-positive composition gives every balance value 0 but is a VALID specimen, scored to the intercept (NOT floored). The score of a row depends only on that row and the frozen model, uses no random numbers and no scored-batch statistics, and is exactly invariant to per-specimen positive scaling.

Usage

score_coda_codacore(model, X, meta = NULL)

Arguments

model: A coda_codacore_model object from fit_coda_codacore.
X: Numeric matrix (samples $\times$ features) of non-negative abundances with named feature columns.
meta: Optional per-sample metadata. Accepted for interface uniformity and ignored by this method.

Value

Plain finite numeric vector of length nrow(X); larger values are more case-like.

References

Gordon-Rodriguez E, Quinn TP, Cunningham JP. (2021) Learning sparse log-ratios for high-throughput sequencing data. Bioinformatics 38(1):157-163.

Examples

if (FALSE) { # \dontrun{
model <- fit_coda_codacore(X, y)
score_coda_codacore(model, X[1, , drop = FALSE])
} # }