Fig. 1: Frozen tissue single nuclei extraction workflow for snDNA-seq.

a Overview of frozen sample single nuclei extraction workflow for Tapestri snDNA-seq. b Representative microscopic images of extracted single nuclei, stained with Trypan blue (brightfield, BF), DAPI (fluorescence, FL). At least 3 representative pictures were taken per sample and yielded similar results. c Technical and genetic profile of each biologically distinct sample. A total of 38 samples were processed, 34 of which were biologically distinct samples from 16 unique patients. Genetic profiles were based on bulk sequencing. d Pseudobulk (p-bulk) variant allele frequency (VAF) comparison of all 160 shared variants between libraries prepared with fresh vs frozen (3 weeks) nuclei of sample PA04-2. Key drivers pre-identified by bulk WES in this case are highlighted; regression line with 90% confidence interval is drawn. e Single-cell genotype (HOM homozygous mutation, HET heterozygous mutation, WT wildtype) heatmap of snDNA-seq libraries generated by fresh vs frozen nuclei of sample PR04-2. Each row represents a bulk data-validated driver variant of this case, while each column represents a single nucleus in the library. The nuclei were sorted based on KRAS variant’s VAF in ascending order from left to right. f Venn diagrams showing colocalization of genotypes belonging to two separate tumor cell populations in one snDNA-seq library (two replicates shown); the cells carrying both genotypes were identified as doublets. Nuclei suspension extracted from two tumor samples from different patients were mixed and subject to snDNA-seq. The two distinct tumor populations were identified by genotype for their respective driver variants TP53 p.C207Y and ARID1A splice. Source data are provided as a Source Data file.