snRNA-seq, WT vs. Gpnmb-perturbed GL261 tumors (Savage et al. 2025)

Summary.
Single nucleus RNA sequencing (snRNA-seq; sci-RNA-seq3 protocol) was used to profile the in vivo brains from mice engrafted with parental or Gpnmb-perturbed GL261 tumors.

Data were preprocessed as described below and only myeloid and tumor cells were retained in the resulting Seurat object.
Data Format.
Data is provided as preprocessed dataset, stored in Seurat Object.

Sample processing, sci-RNA-seq3 library generation, and sequencing
Cells were harvested with 0.25% typsin-EDTA and neuron dissociation solution (Tian et al., 2019), respectively. Cell pellets were immediately snap-frozen in liquid nitrogen and then stored at -80°C for sci-RNA-Seq3-based single-nucleus RNA-Seq processing.

Samples from all conditions were processed together to minimize batch effects. Nuclei extraction and fixation were performed as previously described (Cao et al., 2019), except for the use of a modified CST lysis buffer (Slyper et al., 2020) plus 1% SUPERase In RNase Inhibitor (AM2696). Nuclei quality was checked with DAPI and Wheat Germ Agglutinin (WGA) staining. Sci-RNA-Seq3 libraries were generated as previously described (Cao et al., 2019) using three-level combinatorial indexing. The final libraries were sequenced on an Illumina NovaSeq 6000 using the following protocol: read 1: 34bp, read 2: 69bp, index 1: 10bp, index 2: 10bp.

Raw sequencing reads were first demultiplexed based on i5/i7 PCR barcodes. FASTQ files were then processed using the sci-RNA-Seq3 pipeline (Cao et al., 2019). After barcodes and UMIs were extracted from the read1 FASTQ files, read alignment was performed using the STAR short-read aligner (v2.5.2b) with the mouse genome (hg38) and Gencode v25 gene annotations. After removing duplicate reads based on UMI, barcode, chromosome and alignment position, reads are summarized into a count matrix of M genes x N nuclei.

Filtering
Raw single-cell gene count matrices were loaded into a Seurat object (version 4.0.4) (Butler et al., 2018; Hao et al., 2021; Satija et al., 2015; Stuart et al., 2019) and filtered to retain cells with (i) 200 – 9000 recovered genes per cell, (ii) less than 60% mitochondrial content, and (iii) unmatched rate within 3 median absolute deviations of the median.

Normalization
To normalize expression values, we adopted the modeling framework previously described and implemented in the sctransform R Package (version 0.3.2) (Hafemeister & Satija, 2019). In brief, count data were modelled by regularized negative binomial regression, using sequencing depth as a model covariate to regress out the influence of technical effects, and Pearson residuals were used as the normalized and variance stabilized biological signal for downstream analysis.

Integration
Cells from each treatment condition and differentiation day ere integrated in Seurat using the reciprocal principal component analysis-based approach, using the top 3000 variable features.

Dimensional reduction
PCA was applied to normalized and scaled data, and the top components (accounting for 90% of variance observed in the first 50 PCs) were used for UMAP embedding using RunUMAP(max_components = 2, n_neighbours = 50, min_dist = 01, metric = cosine) in Seurat.

Clustering
To identify clusters, we performed Louvain clustering in Seurat using the FindClusters function.

Contact
Contact Dr. Nicholas Mikolajewicz regarding any questions about the data or analysis ([e-mail])