Uncovering the role of alternative splicing and isoforms in cell fate and disease
Alternative splicing (AS) is a fundamental mechanism of gene regulation that generates multiple isoforms from a single gene, thereby expanding transcript and protein diversity in eukaryotic cells. Isoform expression further refines cell identity and contributes to phenotypic variation across cells, tissues, and individuals. Disruption of normal splicing patterns has been linked to numerous human diseases, highlighting the importance of understanding AS and isoform regulation.
Recent advances in single-cell long-read RNA sequencing enable full-length transcript profiling at single-cell resolution, overcoming limitations of short-read and bulk RNA sequencing by providing both isoform-level resolution and cell-type specificity. Single-cell long-read RNA sequencing has revealed extensive transcript diversity, including many novel isoforms. Despite this potential, the roles of AS and isoforms in cell fate and disease remain underexplored, and dedicated tools for analysing single-cell long-read transcriptomic data remain limited.
The research program at JCU’s Computational BioMedicine lab addresses this gap by developing computational and translational approaches to investigate transcriptomic complexity driven by AS and isoform regulation. These include tools for AS profiling and quantification (SpliceWiz) and quantifying transcript diversity for measuring isoform-level complexity in single cells (ScIsoX). Moreover, we develop machine learning algorithms that predict regulators of AS from multi-omics data or annotate cell types based on isoform expression profiles. We apply these tools and algorithms to analyse aberrant splicing and isoform expression in cancers such as breast cancer and hepatocellular carcinoma. We further investigate fusion gene discovery using long-read sequencing. Collectively, these studies aim to reveal how isoform diversity influences cell fate, improve disease diagnosis and classification, and enable new biological insights.
Poster
ICON: An isoform-aware hierarchical random forest model for cell type classification
Single-cell RNA sequencing (scRNA-seq) has transformed our ability to resolve cellular heterogeneity across complex biological systems. However, conventional short-read scRNA-seq is inherently limited in its inability to capture full-length transcripts. Isoform profiles, arising from alternative splicing, provide a deeper layer of resolution, enabling finer discrimination of cellular subtypes and dynamic states, particularly in heterogenous tissues. Long-read RNA sequencing technologies enable accurate transcript-level profiling and more comprehensive characterisation of isoform diversity. Despite these advances, existing cell type annotation methods remain largely tailored to gene-level data, thereby limiting fidelity and leaving isoform-level information an untapped reservoir of biological insight.
Here, we present a hierarchical random forest framework, ICON, for isoform-aware cell classification in scRNA-seq data. By jointly modelling gene- and isoform-level expression, the framework captures both abundance and useage patterns, enabling classification beyond gene-level resolution. A two-stage strategy first assigns cell identities using highly variable gene and isoform features, followed by targeted reclassification of ambiguous cells based on relative isoform and gene usage, thereby resolving conflicts that arise from transcriptional heterogeneity.
Importantly, ICON provides interpretable outputs by identifying key genes and isoforms that drive cell type discrimination, linking classification to underlying regulatory mechanisms. Benchmarking on long-read scRNA-seq datasets demonstrates consistent improvements over conventional gene-based approaches. With increasing adoption of long-read sequencing, our framework provides a robust, interpretable foundation for isoform-aware cell type annotation, improving resolution and insight.
Bio
Sachini is a final-year MPhil candidate at James Cook University, Townsville, with undergraduate and postgraduate studies in mathematics and computer science, which have provided me with a strong background in mathematical modelling and analysis of large-scale data. My current research focuses on developing an isoform-aware machine learning model for cell type identification using long-read RNA sequencing data. Through my studies, I have developed skills in single-cell long-read transcriptomic analysis, machine learning model development and interpretation, biological data integration, and translating computational outputs into biologically meaningful insights.
