Xenium Leiden Clustering for HistoSeg

Run Locally

Record description

Title: Xenium Leiden Clustering for HistoSeg
DOI: https://doi.org/10.82595/scilifelab.50md5-jq630
Description: Upload a standard 10x Genomics Xenium cell_feature_matrix.h5 file. The app keeps Gene Expression features, performs cell and gene quality control (min_counts and min_genes per cell, min_cells per gene), then runs total-count normalization, log1p, PCA (50 components), a 15-nearest-neighbour graph and Leiden clustering (resolution 1.0, fixed seed). It returns a cluster.csv (Barcode, Cluster; 1-based cluster ids) that can be used directly as input to HistoSeg together with cells.parquet, plus a QC summary with cell/gene counts and cluster sizes. All thresholds and the resolution can be adjusted in the interface.
Subjects and keywords: Transcriptome
Cluster Analysis
Gene Expression Profiling
URL: https://xenium-leiden.serve.scilifelab.se
Source code: https://github.com/longmp1992/xenium-leiden-serve
Docker image: ghcr.io/longmp1992/xenium-leiden-serve:latest
Date submitted: 2026-09-25 11:35
Date updated: 2026-09-25 11:39
Date available: 2026-09-25 11:39
Language: English

Long, Mengping

Version 3:
https://doi.org/10.82595/scilifelab.bq07r-f8m73

Version 2:
https://doi.org/10.82595/scilifelab.jmrsp-p8s37

Version 1:
https://doi.org/10.82595/scilifelab.50md5-jq630 current

To cite all versions, use the following DOI:
https://doi.org/10.82595/scilifelab.w2zew-q0387