Using long-read sequencing to study RNA modifications.
This project examined RNA modification patterns in mouse skeletal muscle using nanopore sequencing. The analysis compared different muscle tissues to identify shared and tissue-associated modification sites.
The broader goal was to understand how RNA modifications may relate to gene regulation and muscle biology, with particular attention to genes relevant to muscular dystrophy and facioscapulohumeral muscular dystrophy (FSHD).
Comparing RNA across muscle tissues.
Nanopore sequencing data were examined across mouse skeletal muscle samples representing different anatomical regions. The analysis included abdominal wall and pectoralis tissue as well as gastrocnemius samples.
Abdominal Wall
AbWall tissue was analyzed as one of the trunk muscle groups in the dataset.
Pectoralis
Pectoralis tissue provided a second trunk muscle dataset for comparison.
Gastrocnemius
Gastrocnemius samples were used to represent limb-associated skeletal muscle.
Looking across multiple RNA modification types.
The analysis examined several classes of RNA modifications detected in the nanopore sequencing data.
Inosine
Inosine sites were analyzed across tissues to identify shared and tissue-associated patterns.
m6A
N6-methyladenosine sites were included as another modification class for comparison.
Ψ
Pseudouridine-associated modification calls were included in the analysis.
m5C / Nm
Additional modification classes including m5C and Nm were examined across the dataset.
From sequencing data to biological interpretation.
The project combined visualization, quantitative analysis, and dimensionality reduction to investigate modification patterns across samples.
Visualizing and comparing modification patterns.
IGV was used to inspect modification sites within genomic regions, while Jupyter-based analysis was used to organize and compare the resulting data. Principal component analysis was used to examine patterns across samples.
IGV
Inspected RNA modification calls at individual genomic locations and examined their distribution across samples.
Jupyter / Python
Used computational notebooks to organize modification data, compare samples, and identify shared patterns.
PCA
Principal component analysis was used to visualize relationships between samples based on modification patterns.
Gene-Level Analysis
Shared modification sites were mapped toward gene-level interpretation and potential relevance to muscle disease.
Narrowing thousands of sites toward biologically relevant genes.
Across the datasets, 8,994 genes were associated with shared inosine sites. Applying a more stringent requirement of at least three shared sites reduced this set to 763 genes.
This filtering step helped narrow the analysis toward genes with repeated modification patterns that could be investigated further in the context of skeletal muscle biology.
Connecting RNA regulation to muscle disease.
RNA modifications provide another layer through which cells can regulate RNA behavior and gene expression. Studying these modifications in skeletal muscle may help identify molecular patterns associated with muscle-specific biology.
The project used computational analysis to move from raw sequencing-derived modification calls toward candidate genes that could be investigated in future studies of muscular dystrophy and FSHD.
Communicating the analysis.
The project was presented as a scientific poster summarizing the sequencing data, computational workflow, modification analysis, and biological interpretation.
View COSMOS Poster →Computational analysis and biological interpretation.
My work involved analyzing nanopore-derived RNA modification data, visualizing modification sites in IGV, working with Jupyter-based analysis, examining sample relationships using PCA, and helping connect computational findings to questions in muscle biology and disease.