Fit REO-UCell within-sample rank discriminator — fit_reo

Learns a frozen up/down feature signature from training data using within-sample ranks, then stores only that signature and fixed hyperparameters for single-sample UCell-style scoring. Training ranks are computed per sample with rank 1 assigned to the most abundant feature. A feature elevated in cases therefore has a lower mean case rank than control rank, and is selected by mean_rank_controls - mean_rank_cases. No test data or test-time batch statistics are used.

Usage

fit_reo_ucell(X_train, y_train, meta_train = NULL, hp = list())

Arguments

X_train: Numeric matrix (samples \(\times\) features) of non-negative abundance values. Columns must be uniquely named feature ids.
y_train: Integer/numeric 0/1 labels aligned to X_train; 1 is case/disease and 0 is control.
meta_train: Optional per-sample metadata. Accepted for interface uniformity and ignored by this method.
hp: List of frozen hyperparameters: k signature size per direction (default min(25L, floor(ncol(X_train) / 4)), at least 1), maxRank UCell rank cap (default min(1500L, ncol(X_train))), and use_down logical (default TRUE). These are resolved once during fitting and are not tuned inside fit_reo_ucell().

Value

A plain list of class reo_ucell_model containing up_features, down_features, maxRank, feature_universe, use_down, k, and the resolved hp.

References

Andreatta M, Carmona SJ. (2021) UCell: Robust and scalable single-cell gene signature scoring. Computational and Structural Biotechnology Journal 19: 3796-3798.

Examples

if (FALSE) { # \dontrun{
set.seed(1)
X <- matrix(stats::rgamma(40 * 30, shape = 2), nrow = 40,
            dimnames = list(NULL, paste0("miR-", seq_len(30))))
y <- rep(c(0, 1), each = 20)
X[y == 1, 1:5] <- X[y == 1, 1:5] + 5
model <- fit_reo_ucell(X, y)
score <- score_reo_ucell(model, X)
} # }