SFMEB

The 'SFMEB' software tool to identify differentially expressed genes in RNA-seq data, considering the complexities of technical effects and varying sequencing depths encountered in experiments. Unlike traditional methods that normalize data using a single scaling factor, 'SFMEB' introduces a scaling-free minimum enclosing ball (SFMEB) method. This approach transforms the problem into an outlier detection scenario, treating differentially expressed (DE) genes as outliers. By constructing the minor possible ball to contain known non-DE genes, the SFMEB method identifies DE genes outside this boundary. Notably, SFMEB does not rely on data normalization and performs well in heterogeneous or biologically replicated settings.