NLSDeconv: an efficient cell-type deconvolution method for spatial transcriptomics data

Abstract

SummarySpatial transcriptomics (ST) allows gene expression profiling within intact tissue samples but lacks single-cell resolution. This necessitates computational deconvolution methods to estimate the contributions of distinct cell types. This paper introduces NLSDeconv, a novel cell-type deconvolution method based on non-negative least squares, along with an accompanying Python package. Benchmarking against 18 existing deconvolution methods on various ST datasets demonstrates NLSDeconvs competitive statistical performance and superior computational efficiency. Availability and implementationNLSDeconv is freely available with tutorial at https://github.com/tinachentc/NLSDeconv as a Python package.