Spatial organization of malignant and immune compartments in HNSCC

An end-to-end Visium analysis across two HNSCC tissue sections.

SPATIAL TRANSCRIPTOMICS · PORTFOLIO PROJECT

How are malignant and immune compartments spatially organized across two HNSCC tissue sections?

This project combines sample-level Seurat analysis, reciprocal PCA integration and SpaCET deconvolution to describe spatial transcriptional domains and sample-relative immune niches in two 10x Genomics Visium sections.

Explore results Review methods Read the scientific report

Scope at a glance

The project contains two biological samples. Cross-sample comparisons are descriptive, spots within each section are not independent biological replicates, and SpaCET fractions are transcriptional estimates rather than exact cell counts.

Project at a glance

2 tissue sections

HNSCC Visium samples 17B5776 and 19H1257, analyzed independently before joint comparison.

4 analysis stages

Quality control, individual clustering, RPCA integration and spatial deconvolution.

Sample-relative niches

Immune-high and Immune-low spots are defined from quartiles calculated separately within each section.

Reproducible outputs

Saved objects, marker tables, SpaCET summaries, spatial figures and a scientific report support the portfolio narrative.

Computational workflow

Import and QC

Load Space Ranger counts and tissue coordinates; inspect spot-level transcript, feature and mitochondrial metrics.

Normalize and cluster

Apply SCTransform, PCA, graph-based clustering and marker-guided spatial-domain assessment within each sample.

Integrate samples

Use a joint representation and RPCA integration to describe shared and sample-associated transcriptional structure.

Resolve compartments

Use SpaCET to estimate malignant, stromal and immune fractions, then map sample-relative immune niches.

Selected findings

Spatial expression of selected lineage- and state-associated markers in 17B5776.

Spatial transcriptional heterogeneity

Observation. Sample 17B5776 contains spatially organized marker programs associated with immune, epithelial, stromal, hypoxic, inflammatory, interferon-responsive and proliferative domains.

Interpretation. These are provisional mixed-spot domain annotations, not pure cell types or formally defined malignant states.

Estimated total immune-fraction distributions across the two HNSCC tissue sections.

Contrasting immune distributions

Observation. The saved SpaCET summary reports a higher median derived immune fraction in 17B5776 than in 19H1257.

Interpretation. This is a descriptive contrast between two sampled sections and cannot be generalized to an HNSCC population.

What remains unresolved

The current workflow characterizes malignant fractions and immune niches, but it does not formally identify distinct malignant transcriptional states. Estimated malignant fractions and epithelial marker patterns should not be presented as a completed malignant-state analysis.

GitHub

Project repository URL is not configured in the current Git checkout. Add the verified repository URL before publication.