Browsing by Author "Adhikari, Sadiksha"

Structural variants comprise a large number of variations occurring in the human genome and are detected in many diseases including cancers. To a limited extent, whole exome sequencing (WES) is capable of detecting structural variations (SVs) using algorithms and tools utilizing local assembly, split-reads, discordant read-pairs and read depth methods. However, due to the significantly large size of SVs compared to the reads produced and the presence of repetitive regions in the genome, identification of SVs presents a major challenge. 10X Genomics has developed a technology that requires very low amounts of DNA and uses a linked-reads approach to produce long reads. Recently, linked-read technology has shown promising results in resolving complex SVs. In this thesis, we aimed to assess whether linked-read exome sequencing is able to infer more comprehensive information in SVs compared to WES in multiple myeloma (MM). The disease model was chosen based on the presence of high numbers of SVs in MM patient tumor cells. Here, we report that linked-read sequencing has led to the identification of a potential novel translocation t(1; 14) that significantly impacts the change in expression of genes and could potentially have impact on the prognosis and treatment of multiple myeloma patients. By Long Ranger analysis we detected t(1;14) in six out of eight samples. Further, to study whether the translocation differentially affects the expression levels of any genes, differential gene expression was performed between t(1;14) positive versus t(1;14) wild type samples. The analysis resulted in 107 differentially expressed genes where 4 upregulated and 103 downregulated genes were found in the translocation positive samples. Among the downregulated genes, we found S100A8 and S100A9 genes which are previously shown to be associated with chemoresistance to PAD (bortezomib, doxorubicin and dexamethasone) therapy. The related breakpoints of the event were identified by Manta tool (SV caller) using both linked-read and WES. Therefore, linked-read information does not appear necessary to detect this event. In this study, we found that linked-read sequencing has certain advantages over WES such as low input DNA, increased number and quality of calls and breakpoint information. However, linked-read sequencing technique is limited to the detection of certain SV types in addition to increased cost of sequencing. These two factors must be considered before choosing linked-read sequencing over WES. Somatic mutations and clinically relevant SV were detected equally efficiently by both techniques.