Point-Process Modelling of Cell-Type Interaction Structure in CosMx Glioma Data for Hypothesis-Driven Tumor Inference from scRNA-seq

Preprint (not peer-reviewed): Carcanholo et al. (Ribeirão Preto Medical School, University of São Paulo) fit whole-tissue Log-Gaussian Cox process models to CosMx data from 8 GBM specimens (2,427,362 cells, 15 cell populations, four malignant states); immune populations show spatial attraction toward malignant states; transcript-based features predict cell-level spatial phenotypes, but leave-one-patient-out validation reduced performance and patient-wise harmonisation did not restore cross-patient transportability. Small cohort (8 specimens); the authors state that larger, more diverse spatial cohorts are needed.

Authors / organisation

Carcanholo FP, Cassiano MHA, Panepucci EM, Laurini MP, Malta TM.

Venue

bioRxiv (preprint, not peer-reviewed)

Date

2026-10-10

Type

preprint

Year

2026

DOI

10.64898/2026.10.02.756019

Verification

bioRxiv API details (title, 5 authors, posted 2026-10-10, v1, new results, CC BY-NC, corresponding institution Ribeirão Preto Medical School - USP, abstract) + Crossref posted-content record fetched 2026-10-20; full text not read

GBMSpatial omicsscRNA-seqImmune

Open in the searchable library → · JSON