Abstract LB-392: Variant analysis of LY6 genes in TCGA ovarian cancer

Abstract

Abstract Background Human Ly6 gene family has been associated with stem cell marker Sca-1 in murine cancer [1]. Sca-1 is known to regulate TGF-b signaling, Wnt signaling and it is important in cancer progression and metastasis in mouse models. Human Ly6 genes are associated with poor clinical outcome in human cancers. Previous studies have shown that this family of genes is highly expressed in Ovarian and Breast cancer compared to normal tissues. Over-expression of these genes was found to be correlated with poor outcome in overall and metastasis free survival. Recent studies have also shown that human Ly6 genes are associated with tumor immune escape and drug resistance [3]. In this poster, we explore the variants in Ly6 and related genes in the TCGA Ovarian Cancer data collection. Materials and Methods We first downloaded RNA seq data of primary tumor tissues from 21 TCGA ovarian cancer patients (in the form of fastq files) from CGHUB (https://cghub.ucsc.edu/), and after quality control, aligned to human reference genome using tool RSEM on the Globus Genomics platform. The BAM file was sorted and PCR duplicates were removed. Variant calling was done on BAM files based on Genome Analysis Toolkit (GATK)'s best practices, to obtain a multi sample variant call file (VCF). After getting the multi sample VCF file, we used SnpEff and Annovar software to annotate and predict the functional effects of variants on genes. SnpSift toolbox was used to filter out variants by extracting only variant of Ly6 genes (and other genes of interest) that passed the quality check, and categorized the output into 4 different groups according to the impact (High, Moderate, Modifier, Low). To validate if these variants were germline or somatic, DNA-seq from a second set of 22 TCGA Ovarian cancer samples (tumor tissue and normal blood samples) were used. BAM files were downloaded and variants were called using the Seven Bridges system. The same filtering steps were applied as above. Results In Set 1, we found variants in CD59, LY6E and LYPD6 mutated in all the 21 cases. We found two stop-loss mutations in CD59 gene, which is responsible for regulating the immune response, tumor cell growth and apoptosis. We found a total of 3794 unique variants short-listed in Set 1, and a total of 8879 unique variants short-listed in Set 2. It was expected to see more variants from the DNA-seq data compared to the RNA-seq data. Among these, 103 unique variants were common to both Set 1 and Set 2. The top results were variants in ESR1, CD44 and LYPD6 family mutated in most samples in both Set 1 and Set 2. We also performed survival analysis on variants in RNA, but not in DNA and found variants in LY6E, PINLYP, LYPD5, ZNF283 significant w.r.t overall survival. Conclusion We found a total of 3794 unique variants (from short-listed set) in Set 1 (TCGA OV RNAseq data), and a total of 8879 unique variants (from short-listed set) in Set 2 (TCGA OV DNAseq). There were total of 103 unique variants common to Set 1 and Set 2 We see evidence of LY6 variants in non-coding regions of RNA (not in DNA) to be significantly associated with overall survival. Citation Format: Krithika Bhuvaneshwar, Midrar Alhossiny, Yuiy Gusev, Subha Madhavan, Geeta Upadhyay. Variant analysis of LY6 genes in TCGA ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr LB-392.