hoodscanR: profiling single-cell neighborhoods in spatial transcriptomics data

Abstract Understanding complex cellular niches and neighborhoods have provided new insights into tissue biology. Thus, accurate neighborhood identification is crucial, yet existing methodologies often struggle to detect informative neighborhoods and generate cell-specific neighborhood profiles. To address these limitations, we developed hoodscanR, a Bioconductor package designed for neighborhood identification and downstream analyses using spatial data. Applying hoodscanR to breast and lung cancer datasets, we showcase its efficacy in conducting detailed neighborhood analyses and identify subtle transcriptional changes in tumor cells from different neighborhoods. Such analyses can help researchers gain valuable insights into disease mechanisms and potential therapeutic targets.