primary ATRT single cell RNA-seq

Seurat object stored in RDS file format. Three fresh tumor samples (respectively from INI254, INI255 and INI267) were prepared and loaded in 10x Chromium instrument (10x Genomics). Libraries were prepared using a Single Cell 3’ Reagent Kit (V2 chemistry, 10X Genomics) and sequenced on an Illumina HiSeq2500 using paired-end 26x98 bp as sequencing mode, targeting at least 50 000 reads par cell. Mapping and UMI counting per gene were performed using cellranger tool (version 3.1.0) and the hg19 reference genome version. Cells with both a low number of genes and a high proportion of mitochondrial RNA were discarded. The threshold of the minimum number of detected genes was set as the 5th percentile of the distribution of the number of detected genes in all cells while the maximum proportion of mitochondrial genes were set by visual inspection of the plot of the number of detected genes versus the percentage of mitochondrial gene of each sample. scRNA-seq data integration was performed using the CCA-based implemented in Seurat version 3. The clustering was conducted using the graph-based modularity optimization Louvain algorithm implemented in Seurat v3. The resolution 0.2 (integrated_snn_res.0.2) was choosen for the final result.