Abstract
Spatially-resolved transcriptomics (SRT) generates large and heterogeneous datasets where global (tissue-wide) quality control (QC) metrics often over-aggressively remove biologically meaningful regions or miss localized artifacts. Recently, spatially-aware QC metrics have been introduced in SpotSweeper, but this is limited to the R programming language, which makes it challenging to use these metrics within the Python/scverse ecosystem. Here, we present SpotSweeper-py, a Python equivalent package of SpotSweeper that computes neighborhood-aware z-scores for standard QC metrics such as total counts, log total counts, number of detected genes, and percentage of mitochondrial counts. We demonstrate the performance and usability of SpotSweeper-py on two public datasets from the 10x Genomics Visium and VisiumHD platforms. This implementation of local spatially-aware QC metrics enables direct integration with Python/scverse ecosystem, reduces false positives from global quality control while preserving tissue-specific architecture. Plotting utilities are also included for quick visualizations of flagged outliers. By making robust local QC accessible in Python, SpotSweeper-py strengthens the reliability of pipelines for analyzing SRT data. The open-source software is available on PyPI (https://pypi.org/project/spotsweeper).