Identification of Novel Transcriptional Alleles in Primary Prostate Cancer Cells and Cancer Stem Cells by Bulk RNA-Seq and Single-Cell RNA-Seq Analyses

Genetic alterations are closely associated with prostate cancer development and progression, but the RNA-derived transcriptional allele landscape of prostate cancer and cancer stem cells (CSCs) remains poorly understood. To characterize cancer-associated transcriptional alleles, we analyzed bulk and single-cell RNA sequencing (RNA-seq) data from primary human prostate cancer cells and their matched benign epithelial cells from non-cancerous regions of the same patients. Sequencing reads were aligned to the human reference genome (hg38) using STAR, and sequence variants were annotated with ANNOVAR. Most detected transcriptional alleles were located in noncoding regions, particularly within 3′ and 5′ untranslated regions, with single-nucleotide variants predominating and C>T substitutions occurring most frequently. Comparison of cancer and matched benign samples identified 223 genes carrying cancer-associated transcriptional alleles consistently detected across three independent patients. Seven of these genes showed differential expression between CSC and non-CSC populations, while 19 contained non-synonymous alleles, including 11 that are predicted to have potentially damaging effects. We identified ATF6 and KDM3A as candidate genes of potential functional interest. Single-cell RNA-seq analyses revealed differences in the number and distribution of detected transcriptional alleles between culture conditions, with 2D cancer cultures detecting more total, coding, and non-synonymous alleles than CSC-enriched 3D spheroids. CSC populations also exhibited fewer detected transcriptional alleles than non-CSC populations. Our findings provide a framework for characterizing transcriptional allele heterogeneity within the prostate cancer cellular hierarchy and identify candidate CSC-associated alterations for further investigation and functional validation. Finally, the cancer–benign matched strategy used in this study provides additional molecular evidence supporting the malignant origin of the tumor-derived cancer cells.