Module arrayweights 

arrayWeights — estimate array quality weights.

Pure-Rust port of limma’s arrayWeights and its internal estimators for the log-linear array-variance model log(sigma_array^2) = Z gamma, with array weights w = exp(-Z2 gamma):

• array_weights — .arrayWeightsREML, the default path (method="auto" with no observation weights and no missing values resolves to "reml"): an exact Fisher-scoring REML similar to statmod::remlscor.
• array_weights_gene_by_gene — .arrayWeightsGeneByGene, a single forward Newton pass used for method="genebygene" (and method="auto" when observation weights or NAs are present).
• array_weights_prwts_reml — .arrayWeightsPrWtsREML, the per-gene REML variant selected by method="reml" when observation weights are present.
• array_weights_quick — arrayWeightsQuick, a fast approximation from an already-fitted model.

Functions

array_weights

Estimate array quality weights by REML.

array_weights_gene_by_gene

Estimate array quality weights by the gene-by-gene update algorithm (.arrayWeightsGeneByGene).

array_weights_prwts_reml

Estimate array quality weights by REML with per-observation prior weights (.arrayWeightsPrWtsREML).

array_weights_quick

arrayWeightsQuick(y, fit): fast approximate array quality weights from an already-fitted model.