xpore: Novel Methods to Identify Differential RNA Modifications

RNA modifications regulate gene expression, stability, and function. Advances in sequencing technologies have revealed their widespread presence, but detecting differences between conditions remains challenging. Traditional methods often require large sample sizes or lack single-molecule resolution, limiting their effectiveness.

New computational approaches are improving detection accuracy. One such method, xpore, leverages nanopore sequencing data to identify differential RNA modifications across biological states.

Major Categories Of RNA Modifications

RNA modifications encompass diverse chemical alterations that fine-tune function, stability, and interactions. Among the most studied is N6-methyladenosine (m6A), which methylates adenosine at the N6 position. This modification influences splicing, translation efficiency, and degradation, playing roles in embryonic development and neuronal function. METTL3 and METTL14 catalyze m6A deposition, while demethylases like FTO and ALKBH5 remove these marks, creating a dynamic regulatory system. Studies using m6A-seq and nanopore direct RNA sequencing show m6A enrichment in 3′ untranslated regions (UTRs) and near stop codons, where it modulates mRNA stability and translation.

Other methylation-based modifications also contribute to RNA regulation. 5-methylcytosine (m5C) is found in mRNA and non-coding RNAs, including tRNAs and rRNAs, where it affects stability and translation. NSUN family methyltransferases catalyze m5C, and its presence in mRNA has been linked to stress responses and cancer. N1-methyladenosine (m1A) appears in coding and non-coding RNAs, influencing translation initiation and ribosome interactions. Unlike m6A, which is found internally within mRNA, m1A is enriched near the 5′ UTR, enhancing translation under specific conditions.

Pseudouridine (Ψ), the most abundant RNA modification, results from uridine isomerization, creating a more stable nucleotide. It is prevalent in ribosomal RNA (rRNA) and transfer RNA (tRNA), enhancing ribosome function and translation fidelity. Pseudouridine also appears in mRNA, where it alters codon-anticodon interactions and influences stability. Studies using pseudouridine-seq show Ψ levels fluctuate in response to cellular stress, suggesting an adaptive role. Similarly, inosine (I), generated by adenosine-to-inosine (A-to-I) editing via ADAR enzymes, alters base-pairing properties and expands RNA coding potential. This modification is particularly important in neural tissues, where it contributes to transcriptome diversity by modifying neurotransmitter receptors and ion channels.

Key Factors Influencing Modification Patterns

RNA modifications are shaped by enzymatic activity, cellular conditions, and external influences. Their distribution is tightly regulated by methyltransferases, demethylases, and other modifying enzymes. For instance, m6A deposition, controlled by the METTL3-METTL14 complex, preferentially targets conserved sequence motifs near stop codons and 3′ UTRs. RNA-binding proteins, such as YTH domain-containing proteins, recognize modified sites and influence RNA fate. The balance between methylation and demethylation is dynamic, allowing reversibility in response to physiological demands.

RNA secondary structures also determine modification sites. Pseudouridine and m5C are enriched in structured regions where they enhance stability and protein interactions. Structural constraints can promote or hinder enzyme accessibility, leading to selective modification patterns. For example, m1A is often found in stem-loop structures where it enhances translation efficiency, particularly under stress. Similarly, A-to-I editing by ADAR enzymes occurs preferentially in double-stranded RNA regions, altering base-pairing properties and expanding transcriptome diversity. This interplay between RNA folding and modification dynamics influences splicing, localization, and degradation.

Environmental factors and cellular stressors contribute to modification variability. Hypoxia, oxidative stress, and nutrient availability alter RNA modification landscapes, affecting gene expression programs. Under heat shock, m6A levels increase on specific transcripts, facilitating rapid protein synthesis. Oxidative stress induces changes in m5C and pseudouridine patterns, modulating RNA stability and translation. In cancer, aberrant modification patterns enhance oncogene expression or silence tumor suppressors, contributing to tumor progression.

xpore For Differential RNA Modification Detection

Detecting differential RNA modifications has been challenging due to limitations in traditional sequencing methods, which often require large sample sizes or indirect chemical treatments. xpore, a computational framework leveraging direct nanopore sequencing data, offers a refined approach by identifying changes in RNA modification patterns at the single-molecule level. Unlike antibody-based or bisulfite sequencing methods, which may introduce biases or miss subtle modifications, xpore uses unique electrical signal variations from modified nucleotides passing through a nanopore. These signal deviations allow direct modification detection without prior enrichment or chemical conversion, providing a high-resolution view of RNA modifications.

At xpore’s core is a statistical model comparing nanopore current signal distributions between experimental and control conditions. By analyzing base-calling errors and current shifts, xpore infers modifications and determines whether their levels differ significantly. This approach is particularly useful for studying dynamic RNA modifications that fluctuate in response to physiological changes, enabling transcript-specific tracking. Additionally, xpore employs a Bayesian framework to quantify uncertainty, reducing false positives and increasing confidence in differential modification detection.

A major advantage of xpore is its ability to analyze full-length RNA molecules, preserving modification context within entire transcripts. This feature is crucial for understanding how modifications influence RNA processing events such as alternative splicing, stability, and translation. Applying xpore to nanopore sequencing data from human cell lines has revealed condition-specific m6A alterations correlating with differential gene expression. The tool’s adaptability extends to various RNA species, including long non-coding RNAs and circular RNAs, which are often difficult to analyze using short-read sequencing platforms. By maintaining sequence integrity and modification context, xpore provides a comprehensive view of post-transcriptional regulation.

Physiological And Cellular Significance

RNA modifications influence gene expression, ribosome activity, and metabolic adaptation. Their presence in mRNA, rRNA, and tRNA enables fine-tuned control over protein synthesis, allowing cells to adjust translation efficiency in response to environmental and developmental cues. Modifications such as m6A regulate transcript stability and translation rates, facilitating dynamic gene expression changes necessary for cellular differentiation and tissue development. In embryonic stem cells, m6A modifications modulate the degradation of key transcription factors, ensuring temporal and spatial gene expression control during development.

Beyond embryogenesis, RNA modifications contribute to neuronal plasticity by shaping transcript stability and localization within synapses. Pseudouridine and A-to-I editing in neural transcripts affect neurotransmitter receptor function, influencing synaptic strength and cognition. Disruptions in these modifications are linked to neurodevelopmental disorders such as autism and schizophrenia. Additionally, modifications within rRNA and tRNA influence ribosome assembly and translational fidelity. Cells experiencing metabolic stress exhibit dynamic shifts in tRNA modifications, altering codon usage preferences and promoting the translation of stress-responsive proteins. This adaptive mechanism enhances cellular resilience by prioritizing essential protein production under adverse conditions.

Links To Genetic Variation

RNA modification patterns are influenced by genetic variation, which can affect modification efficiency and downstream gene expression. Single-nucleotide polymorphisms (SNPs) near modification sites can alter methyltransferase and demethylase activity, leading to individual differences in modification levels. For example, genetic variants in METTL3 and METTL14 have been linked to altered m6A deposition, affecting mRNA stability and translation. Genome-wide association studies (GWAS) have identified SNPs in RNA modification-related genes correlated with disease susceptibility, highlighting the impact of genetic variation on post-transcriptional regulation.

Sequence polymorphisms can also influence RNA secondary structure, affecting modification accessibility. Certain mutations create or disrupt structural motifs that guide modification enzymes, altering transcript regulation. This is particularly evident in A-to-I editing, where sequence variants in double-stranded RNA regions enhance or reduce editing efficiency. Such variations have been implicated in neurological disorders, as A-to-I editing plays a critical role in neural transcriptome diversity. Additionally, inherited mutations in RNA-modifying enzymes have been linked to diseases such as dyskeratosis congenita, which is caused by defects in pseudouridine synthesis, leading to impaired ribosome function. As sequencing technologies advance, integrating genetic variation data with RNA modification profiling will provide deeper insights into how inherited and somatic mutations shape post-transcriptional regulation, influencing phenotypic diversity and disease risk.