Spectral Compression of Single-Cell Transcriptomes. A Proof-of-Concept FFT Framework for Scalable MRD Follow-updocx

Abstract Single-cell RNA sequencing (scRNA-seq) enables high-resolution profiling of cellular heterogeneity, yet its high dimensionality and sparsity remain major challenges for downstream analysis and visualization. This study introduces a framework that applies the Fast Fourier Transform (FFT) to scRNA-seq expression profiles, converting gene-level signals into frequency-domain representations. By selectively retaining low-frequency components, the method reduces dimensionality while preserving biologically meaningful structure and suppressing technical noise. The approach facilitates clustering, visualization, and annotation tasks. Applied to the PBMC10k dataset, FFT-based compression achieved performance comparable to conventional methods while offering modularity, transparency, and computational efficiency. This framework provides an interpretable and biologically grounded preprocessing strategy for scalable single-cell analysis across transcriptomic and other omics modalities