20.500.12556/DiRROS-25414 Learning tissue representation by identification of persistent local patterns in spatial omics data Spatial omics data provide rich molecular and structural information on tissues. Their analysis provides insights into local heterogeneity of tissues and holds promise to improve patient stratification by associating clinical observations with refined tissue representations. We introduce Kasumi, a method for identifying spatially localized neighborhood patterns of intra- and intercellular relationships that are persistent across samples and conditions. The tissue representation based on these patterns can facilitate translational tasks, as we show for stratification of cancer patients for disease progression and response to treatment using data from different experimental platforms. On these tasks, Kasumi outperforms related approaches and offers explanations of spatial coordination and relationships at the cell-type or marker level. We show that persistent patterns comprise regions of different sizes, and that non-abundant, localized relationships in the tissue are strongly associated with unfavorable outcomes. microenvironment tumour heterogeneity Kasumi mikrookolje heterogenost tumorjev Kasumi true false true Nature Publishing Group Združeno kraljestvo Angleški jezik Slovenski jezik © The Author(s) 2025 Neznano 2026-01-20 09:34:25 2026-01-20 09:34:26 2026-01-21 03:49:59 0000-00-00 00:00:00 2025 0 0 Nasl. z nasl. zaslona; Soavtor iz Slovenije: Jovan Tanevski; Opis vira z dne 19. 1. 2026; str. 1-15 Vol. 16, article no. 4071 30 Apr. 2025 0000-00-00 Zaloznikova Objavljeno NiDoloceno 2024-03-26 2025-04-23 2025-04-30 004.8 2041-1723 10.1038/s41467-025-59448-0 265161475 https://www.nature.com/articles/s41467-025-59448-0 1 d6dc5d74-f5da-11f0-a15d-001a4af901a5 https://dirros.openscience.si/Dokument.php?lang=slv&id=38486 RAZ_Tanevski_Jovan_2025.pdf RAZ_Tanevski_Jovan_2025.pdf 1 E97DB355194BE084BE2B150007D5BC03 bb0a1c5ea3a3d26459110291010badb9c141a73d1f786fb7c1e9667ace4a5c9d 30197485-f5db-11f0-a15d-001a4af901a5 https://dirros.openscience.si/Dokument.php?lang=slv&id=38487 Institut Jožef Stefan 0 0 0