Predict cross-technology harmonized rCLR scores
Usage
predict_cross_tech_harmonized_rclr(
fit,
expr,
meta,
technology_col = fit$technology_col,
min_anchors_present = 3L
)
Arguments
- fit
Fit from `fit_cross_tech_harmonized_rclr()`.
- expr
Numeric matrix, samples x features.
- meta
Data frame aligned to `expr`; technology labels are validated but
not used to select anchors.
- technology_col
Column identifying technology class.
- min_anchors_present
Minimum anchors required in `expr`.
Value
Numeric score vector with attribute `n_anchors_present`.
Details
Insufficient frozen anchor support raises a singlesample_input_error with code INSUFFICIENT_CROSS_TECH_ANCHORS, instead of an all-NA score vector.