WYMM Tour: Seoul

Long-read RNA sequencing has the capability to resolve previously undetected splicing isoforms. In this study, we employed Oxford Nanopore Technologies to obtain long-read transcriptome data from prostate and renal tumor cell lines. Our analysis identified approximately 22,000 distinct isoforms, including around 3,000 candidates for novel splicing isoforms, novel fusion genes, or novel genes. Among these transcripts, we found four fusion gene transcripts exhibiting misregulated transcription in the tumor cell lines, where adjacent genes were transcribed as a single transcript. Additionally, we identified a novel candidate isoform of FOLH-1, the prostate-specific membrane antigen gene. These findings provide valuable splicing resources for understanding the cancer transcriptome.

The identification of genetic causes serves as a pivotal starting point in patient care for rare diseases. With the advent of next-generation sequencing, many patients with rare conditions have undergone genetic diagnoses, leading to the discovery of numerous disease-causing genes. However, despite undergoing various genetic tests, including whole-exome or whole-genome sequencing, a substantial number of patients still lack a clear genetic diagnosis. Long-read sequencing technology introduces new potential for identifying a diverse range of genetic mutations that were previously inaccessible through conventional short-read sequencing. This technology enables the reading of significantly longer segments of DNA or RNA molecules, aiding in deciphering complex genetic structures and identifying mutations within lengthy nucleotide sequences. These advancements open up new diagnostic possibilities for patients with rare diseases, revealing genetic causes that were previously elusive. In this presentation, I aim to briefly introduce the history of clinical genetics in the aspect of diagnostics and share how recent advancements in genomics will broaden the horizons of genetic diagnostics for patients with rare diseases.

Nanopore sequencing by Oxford Nanopore Technologies is a long-read sequencing method that has considerable advantages for characterising mRNA isoforms. It works by recording changes in electrical current when a DNA or RNA molecule traverses through a pore. However, basecalling of this raw signal (known as a squiggle) is error prone, which makes it challenging to perform tasks such as RNA splice site identification or cell barcode identification in single-cell nanopore RNA sequencing data. Existing strategies typically rely on using matched short-read data and/or annotations, which can increase costs or restrict applicability in many situations. In this talk, I will introduce the computational methods my team has developed to perform these tasks solely using the nanopore data: NanoSplicer for splice junction identification and BLAZE for cell barcode identification. I will illustrate the advantages of these methods through data analysis. Finally, I will discuss several other challenges in the field of long-read RNA-seq analysis.

Large scale genomic studies are cataloguing thousands of genetic mutations. With the sheer number of discovered mutations, determining their phenotype and functional characterization remains an enormous challenge. Conventional CRISPR screening have been used to determine the phenotypic effects of genetic mutations by analyzing altered cellular fitness. However, these methods were not able to determine CRISPR perturbed cells’ state in detail. Single-cell CRISPR screening enabled transcriptome changes induced by CRISPR engineering. However, they had relied on short-read sequencing which are not able to detect full-length transcripts. Herein, we developed novel single-cell technologies to directly introduce mutations into the human genome and determine their transcriptional phenotype with integrated long- and short-read sequencing among individual cells. These enable introduction of more diverse genome engineering (i.e., gene KO, point mutations, gene fusions, etc.) and investigation of their effect in depth (i.e., genetic mutations, transcript isoform usage, etc.) in a single-cell resolution. This will generate rich and valuable dataset about complex phenotypes of various genetic mutations which were not attainable before with previous methods.

Aquilaria sinensis, or incense tree, is a plant species belonging to the Thymelaeaceae family. A. sinensis is an evergreen tree with a height of 6-20m and has been used as a source of fragrance or medicinal purposes. In this study, we built a genomic architecture of A. sinensis, by integrating Nanopore, Illumina, and Pore-C sequencing data. The final genome size was ~846.6Mb, with a total contig number of 138 and a contig N50 of 34.3Mb. We combined the Pore-C data to generate chromosome level of scaffolds. Eight super scaffolds were generated indicating 2n=16, with scaffold N50 of 100.8Mb. BUSCO evaluation revealed that the genome completeness reached 96.8%. The repeat sequences accounted for 31.3% mainly by retroelements. A total 36,176 protein-coding genes were annotated from A. sinensis genome. The additional analyses for genome characterization is underway.

RNA regulation is a complex process that ensures precise gene expression, relying on the intricate interplay between RNA and RNA-binding proteins (RBPs). To understand how RBPs collectively influence gene expression, it is essential to obtain detailed, dynamic views of these interactions at the single-molecule level. We have developed a novel full-length RBP footprinting technique that combines in vivo chemical cross-linking, nanopore direct RNA sequencing (DRS), and machine learning methods. This new approach enables the detection of RNA-binding events along entire RNA molecules, capturing a pseudo-temporal sequence of interactions throughout the RNA lifecycle. By examining the patterns of RBP footprints across various cellular RNAs, we aim to identify and analyze the dynamic networks of interactions that evolve over the lifespan of an RNA molecule. Our preliminary results demonstrate the potential of our method to investigate specific RBP interactions, such as the step-by-step assembly of the telomerase complex and the dynamics of the exon-junction complex formation and disassembly. We plan to extend this method to a broad array of cellular RNAs to reveal how RBP bindings are structured into functional modules that govern cellular processes. This comprehensive approach will enable us to examine how RBPs modify the transcriptome and translatome, ultimately reshaping cellular functions, to gain in-depth insights into the complex interplay between RNAs and RBPs.

In this presentation, we will discuss the use of the Nanopore Flongle platform in microbial genome analysis. The Nanopore Flongle is a powerful sequencing tool that offers researchers the opportunity to conduct various studies at a low cost. However, depending on the research objectives and content, there are still issues that need to be resolved. In this presentation, we will explore the potential of Nanopore Flongle in microbial genome research by introducing various research examples. We will also discuss the benefits and limitations encountered during the research process.

Salmonella is one of the food-borne pathogens that causes major public health problems. Salmonella consists of only two major species. However, Serotypes are further classified into more than 2,500 serovars based on their somatic (O) and flagellar (H) antigens. Serotyping of Salmonella spp. is commonly performed through using the Kauffmann–White scheme, but it is too time consuming and less accurate. Therefore, the development of a new serotyping method is necessary. In this study, we confirmed a Salmonella serotype using Oxford Nanopore Technology protocol. The protocol contains sample preparation, library preparation, and data analysis which is enabling the identification of approximately 2,500 serovars. To evaluate this, we tested 10 strains, including Enteritidis and Typhimurium which are frequently found globally. The NGS analysis results confirmed the detection of all 10 species. The serotyping method using the NGS Panel has been confirmed to detect serotypes with 100% accuracy. It is thought that the newly developed NGS panel can be used to quickly and accurately screen for Salmonella serotypying.