rqubic

"rqubic" introduces a biclustering algorithm, QUalitative BIClustering (QUBIC), which significantly advances the analysis of gene expression data by identifying subgroups of genes that exhibit similar expression patterns under specific subsets of experimental conditions. Biclustering, an extension of traditional clustering techniques, has shown potential in unraveling complex biological relationships but has yet to be utilized due to the lack of efficient and reliable algorithms for addressing the general biclustering challenge.

QUBIC distinguishes itself by employing qualitative (or semi-quantitative) measures of gene expression data and a combinatorial optimization technique, enabling it to tackle the biclustering problem more broadly than existing algorithms. Notably, QUBIC can identify all statistically significant biclusters, including those with 'scaling patterns'—a particularly challenging aspect of biclustering previously considered difficult to resolve. Furthermore, QUBIC's efficiency is unparalleled; it can process biclustering problems involving tens of thousands of genes and up to thousands of conditions within minutes on a standard desktop computer.