More than a Scaffold: The Evolving Role of Ribosomal RNA (rRNA) in Translational Regulation

Meta intent: This article reframes ribosomal RNA as both a regulatory layer in translation and the dominant technical bottleneck in precision RNA-seq.

Ribosomal RNA has long been trapped in two incomplete definitions. In basic molecular biology, it is introduced as the structural core of the ribosome. In sequencing workflows, it is treated as background that must be removed before useful transcriptome data can emerge. Both statements are true. Neither is enough. In 2026, rRNA needs to be understood in a broader and more practical way. It is not just a scaffold. It is not just a nuisance. It is a chemically active component of the translational machine, a plausible source of ribosome heterogeneity, and the single largest factor that can flatten sensitivity in total RNA sequencing if library design is not handled well.

That dual role changes how RNA projects should be planned. If rRNA is only noise, the answer is simple: remove it as efficiently as possible. If rRNA also carries regulatory information, the design problem becomes more selective. Some projects should deplete it. Some should preserve it. Some should profile it directly. The right answer depends on the question, not on habit. This is where many generic rRNA articles fall short. They define the molecule, but they do not explain when rRNA should be treated as background, when it should be treated as signal, and when it should be treated as both.

A useful rRNA resource therefore has to do three jobs at once. It has to explain the architecture of the ribosome. It has to explain why ribosomes may not be chemically identical across all contexts. And it has to explain why rRNA depletion remains the decisive step in many RNA-seq workflows. Those are not separate stories. They are different views of the same molecule. The first is structural. The second is regulatory. The third is technical. Together, they define the modern ribosomal paradigm.

The Architecture of the Ribosomal Machine

The ribosome remains one of the clearest demonstrations that RNA is more than an information carrier. It is also part of the machinery that executes gene expression. In bacteria, the canonical ribosome is described as 70S, composed of a 30S small subunit and a 50S large subunit. In the eukaryotic cytosol, the canonical ribosome is 80S, composed of a 40S small subunit and a 60S large subunit. These values are sedimentation coefficients rather than additive masses, but they still capture real differences in architecture and composition.

The small subunit is responsible for decoding. It engages the mRNA, monitors codon-anticodon pairing, and helps position incoming tRNAs. The large subunit performs peptide bond formation and coordinates the chemistry of elongation. This division of labor is familiar. What matters more is what enables it. The ribosome is not a protein shell that happens to contain RNA. It is a ribonucleoprotein machine in which rRNA defines much of the geometry, spatial organization, and catalytic environment required for translation.

In eukaryotes, the 40S subunit contains 18S rRNA and its associated ribosomal proteins. The 60S subunit contains 28S, 5.8S, and 5S rRNAs plus its own protein complement. In bacteria, the 30S subunit contains 16S rRNA, while the 50S subunit contains 23S and 5S rRNAs. These facts still matter because they map directly onto experimental use cases. A mammalian project focused on translational regulation, a bacterial expression study, and a microbial community survey are all rRNA-centered in very different ways.

The ribosome is also one of the strongest examples of RNA-based catalysis. The peptidyl transferase center is not a classic protein enzyme pocket. It is an RNA-rich catalytic environment inside the large subunit. That detail changes how rRNA should be described. A molecule that defines catalytic geometry is not passive support. It is part of the mechanism. That is why the phrase “more than a scaffold” is not rhetorical. It is mechanistic.

Figure 1. High-resolution 3D ribosome structure showing rRNA, tRNA, and mRNA interactions. The figure distinguishes the 40S and 60S subunits, highlights the major rRNA components, traces the mRNA path through the decoding channel, shows tRNA occupancy in the A, P, and E sites, and visually emphasizes the RNA-rich peptidyl transferase center.

This structural view also explains why rRNA stays relevant across so many areas of life science. In molecular biology, it is central because the ribosome cannot function without it. In sequencing, it is central because its abundance overwhelms many libraries. In phylogenomics, it is central because conserved and variable regions of rRNA genes make them powerful classification markers. The same molecule becomes a catalytic core, a technical obstacle, or an evolutionary anchor depending on the design of the experiment.

That breadth is exactly why a flat rRNA overview no longer works. A useful 2026 article has to move quickly from ribosomal architecture to the harder question: is rRNA simply required for translation, or can it also tune translation in a condition-specific way? Once that question is taken seriously, the discussion has to move beyond static structure into ribosomal heterogeneity.

Ribosomal Heterogeneity and the Specialized Ribosome Hypothesis

For many years, the default assumption was that ribosomes inside a cell were functionally interchangeable. They might vary slightly in assembly state or ribosomal protein occupancy, but the core expectation was uniformity. That assumption was reasonable. Ribosomes are deeply conserved, and translation is too essential to tolerate obvious instability. But the field has shifted. The specialized ribosome hypothesis proposes that not all ribosomes are chemically identical, and that those differences may influence translational output.

The logic is simple. If ribosomes differ in composition, and if those differences alter decoding behavior, elongation dynamics, factor interactions, or transcript preference, then ribosomes may not treat every mRNA the same way. In that case, the ribosome stops being a generic machine and becomes part of the regulatory logic of gene expression.

Some heterogeneity may come from ribosomal protein composition. Some may come from assembly history or subcellular localization. Some may come directly from the rRNA itself. That last category is the most important here. Sequence variation, modification state, and local structural differences in rRNA can all create chemically distinct ribosome populations. Those populations do not need to behave in dramatically different ways to matter. Small translational biases can produce large phenotypic effects when the affected targets encode pathway regulators, developmental switches, or stress-response factors.

This idea also helps explain a recurring observation: transcript abundance and protein abundance often do not align cleanly. Many explanations can account for that mismatch, including RNA localization, differential decay, translational repression, and protein turnover. But ribosome chemistry introduces another layer. If certain transcripts encounter ribosomes with one rRNA state while others encounter ribosomes with another, translation can become selective even without large changes in mRNA level.

That does not mean every divergence between RNA and protein should be attributed to specialized ribosomes. It means the concept now deserves a place in experimental reasoning, especially in systems where subtle control matters. Development, stress adaptation, transformed cellular states, and controlled perturbation models are all settings where modest translation biases can reshape phenotype.

Tissue-Specific rRNA Isoforms and Sequence Variation

One possible source of heterogeneity lies in the rRNA sequence itself. This is harder to analyze than it sounds. rRNA genes are repetitive, highly similar, and historically difficult to resolve in standard pipelines. That technical difficulty has often pushed sequence-level variation to the edge of the conversation. But difficult does not mean irrelevant. If different tissues, developmental states, or perturbation conditions express different rRNA variants, even modest sequence changes could alter folding kinetics, ribosomal protein interaction surfaces, or local decoding geometry.

This question becomes especially relevant when translation needs to be selective rather than merely active. A proliferating cell, a stressed neuron, and an activated immune cell do not necessarily benefit from identical translational priorities. If rRNA sequence diversity contributes to those priorities, then the ribosome becomes part of regulatory identity rather than a universal chassis.

Short-read RNA-seq is not always ideal for this problem. It excels at abundance measurement and large-scale comparison, but it is weaker when repetitive loci, full-length context, or combined sequence-plus-modification states matter. That is why native-molecule strategies that preserve full-length RNA context can become more informative than conventional short-read workflows in rRNA-centered studies. For projects that need direct access to intact RNA molecules, Nanopore Direct RNA Sequencing provides a more suitable entry point than assembly-dependent inference.

Epitranscriptomic Signatures: Pseudouridylation and 2′-O-Methylation

The strongest current evidence for ribosome heterogeneity comes from rRNA modification biology. Two classes dominate the discussion: pseudouridylation and 2′-O-methylation. Both are abundant in rRNA. Both alter local RNA chemistry. And both can influence how the ribosome behaves.

Pseudouridine can stabilize local structure through altered hydrogen bonding and base stacking. 2′-O-methylation changes ribose chemistry and can affect local rigidity, conformation, and interaction surfaces. These are not decorative marks. In a highly folded RNA machine, local chemical changes can alter the physical behavior of the fold. Once the fold changes, local function can change with it.

The key advance is not only that these marks exist. It is that they can vary. An rRNA position does not need to be fully modified in every ribosome. Partial stoichiometry is enough to create mixed ribosome populations. One condition may generate a ribosome pool in which a given site is heavily modified. Another may reduce that fraction. The sequence remains the same, but the ribosome population becomes chemically heterogeneous.

That matters because translation control often works through bias rather than absolute switching. If a ribosome subpopulation becomes better or worse at supporting translation of a certain transcript class, the phenotype may change without dramatic reprogramming. What is well supported is that fractional modification exists. What remains harder is assigning transcript-selective consequences at scale in every system. That distinction matters. The field has moved beyond asking whether rRNA modifications are present. It is still working out when a particular modification change can be linked confidently to selective translation in a specific biological context.

This has an immediate workflow consequence. In a standard expression study, aggressive rRNA removal is often correct because the goal is to recover the rest of the transcriptome. In a ribosome-state study, complete removal can erase the very signal of interest. That is why the project question has to come first. A discovery-stage design may begin with RNA-Seq to map overall transcript shifts. If the biology then points toward translational control, a translation-state assay such as Ribosome Profiling (Ribo-seq) or Polysome Sequencing becomes the more informative next step. If the study finally needs to examine chemical variation within native rRNA itself, the Nanopore RNA Methylation Sequencing Service or the 2′-O-RNA Methylation Sequencing Service becomes a more targeted choice.

Technical Sovereignty: Why rRNA Dominates RNA-Seq Design

The biological importance of rRNA is only half the story. The other half is practical. In many total RNA preparations, rRNA accounts for the overwhelming majority of molecules. If it is left in the sample, it can consume most sequencing reads and flatten the effective complexity of the library. This is the core of the “90% problem.” The problem is not only waste. It is loss of sensitivity. Every read spent on abundant ribosomal background is a read not spent on low-copy transcripts, unstable intermediates, rare lncRNAs, or narrow condition-specific signals.

This is why rRNA handling is one of the most consequential choices in RNA-seq design. Researchers often focus on platform, depth, and downstream analysis. Those factors matter. But none of them can rescue a library that has already lost informative complexity during sample preparation. The sequencer can only measure what remains after the library is built.

Figure 2. RNA quality traces before and after rRNA depletion. The figure displays a Bioanalyzer-style comparison in which dominant ribosomal peaks occupy most of the signal before depletion and are greatly reduced after depletion, allowing broader visibility of informative transcript regions.

The common decision is framed as rRNA depletion versus mRNA enrichment. That is useful, but incomplete. Poly(A) enrichment narrows the library to polyadenylated transcripts while reducing ribosomal background indirectly. rRNA depletion tries to subtract the dominant ribosomal fraction while preserving a broader RNA population. These methods are not interchangeable. They answer different biological questions and tolerate different kinds of failure.

How to Decide Whether rRNA Should Be Removed, Preserved, or Profiled Directly

A practical rRNA study design question is not simply how to remove ribosomal background. It is whether the project benefits more from subtraction, preservation, or direct profiling of rRNA-associated states. That distinction is where many workflow discussions remain too shallow.

If the project is a broad expression study and the main concern is sensitivity for low-abundance transcripts, then removing rRNA is usually the correct choice. The goal here is not to learn from rRNA itself. The goal is to prevent abundant structural RNA from dominating the library. Depletion-based total RNA workflows are especially useful when the researcher wants broad transcript retention without losing non-polyadenylated RNA classes.

If the project aims to recover a wide RNA landscape that includes non-coding RNA, histone mRNA, or precursor transcripts, then preserving non-rRNA diversity while subtracting the ribosomal fraction becomes the real objective. In that case, the study benefits less from narrow mRNA enrichment and more from broad-retention designs such as Total RNA Sequencing, where the biological value comes from keeping the transcript universe broad even after rRNA subtraction.

If the project asks whether ribosome state, modification state, or translational selectivity has changed, then directly profiling rRNA becomes more valuable than simply removing it. In those settings, the ribosome is no longer background. It becomes part of the biological signal. That is where translation-state assays and native-molecule methods become more informative than conventional depletion alone. For studies centered on translational occupancy, Ribosome Profiling (Ribo-seq) or Polysome Sequencing can reveal which messages are actually engaging ribosomes. For studies centered on native rRNA chemistry, Nanopore Direct RNA Sequencing provides the intact-molecule context needed to ask modification-aware questions.

Sample type also changes the decision. Intact mammalian RNA behaves very differently from degraded input, bacterial RNA, or mixed host-pathogen material. If the sample is degraded, poly(A) capture may narrow the useful signal too aggressively. If the sample is taxonomically mixed, sequence compatibility and depletion chemistry become harder to predict. That is why the correct workflow often depends less on vendor branding than on the biological predictability of the RNA background.

The right habit, then, is question-first design. Ask what must remain visible after library preparation. Ask what type of distortion is safest. Ask whether rRNA is the problem, the context, or the analyte. Only then choose the method.

The Physics of rRNA Depletion

rRNA depletion is often described as a kit choice. In reality, it is a physical process built on target recognition, hybridization, molecular accessibility, and selective removal. Every one of those steps can fail for a reason rooted in molecular behavior rather than protocol wording.

The first variable is sequence complementarity. Probe-based depletion depends on the ability of an oligonucleotide to find and bind its intended rRNA target. When the target sequence is predictable, that can work very well. But the problem becomes harder in non-model organisms, microbial consortia, environmental RNA, and host-pathogen mixtures. Even a strong probe set can lose efficiency if the actual rRNA sequence landscape deviates from the design assumption.

The second variable is accessibility. rRNA is not a loose linear strand. It is highly folded. Some regions are accessible. Others are buried within stable secondary and tertiary structure. Hybridization therefore depends not only on sequence match, but also on whether the target region is physically exposed under the reaction conditions.

The third variable is input integrity. Fragmented RNA behaves differently from intact RNA. In some cases, fragmentation exposes buried target regions and improves access. In other cases, it destroys the continuity needed for stable hybridization or efficient capture. Degraded RNA is not simply weaker intact RNA. It is a different molecular substrate.

The fourth variable is transcript preservation. Every depletion workflow is supposed to remove rRNA while leaving the rest of the transcriptome intact. In practice, that boundary is imperfect. Off-target binding, unintended capture, and reaction-adjacent bias can all reshape the remaining RNA pool. That becomes especially important when the downstream study depends on low-copy regulatory RNAs, unstable intermediates, or weak signals.

Figure 3. Side-by-side mechanism diagram of RNase H-mediated depletion and bead-based subtractive hybridization. The left side illustrates DNA probe binding, RNA:DNA duplex formation, and RNase H cleavage. The right side displays capture-probe hybridization, bead immobilization, and physical removal of rRNA, highlighting the physics of depletion.

This is why depletion quality should be treated as an experimental variable, not an invisible preprocessing step. The question is not only whether rRNA was reduced. The more useful question is whether it was reduced consistently, specifically, and without distorting the RNA classes that the study actually depends on.

RNase H-Mediated Cleavage vs. Bead-Based Subtractive Hybridization

Two depletion logics dominate most practical workflows: RNase H-mediated cleavage and bead-based subtractive hybridization. They aim at the same target. They do not remove it in the same way.

In RNase H-mediated depletion, short DNA oligonucleotides are designed to hybridize to rRNA. Once an RNA:DNA duplex forms, RNase H cleaves the RNA strand within that hybrid. The unwanted ribosomal RNA is therefore fragmented by enzymatic action. When probe design is strong and the target sequence is well characterized, this can be efficient and adaptable.

But RNase H workflows are not universally robust. They depend on good probe-target matching. They depend on access to hybridization sites. They depend on reaction conditions that allow cleavage across a heterogeneous input population. If some regions are easier to hybridize than others, depletion can become uneven. If the RNA is degraded, the target landscape changes again. The output may still improve, but the depletion pattern may become patchy rather than uniform.

In bead-based subtractive hybridization, the unwanted rRNA is not cleaved first. It is captured. Probes bind to rRNA molecules, and the hybrids are then immobilized on beads or another solid support. The bead-bound fraction is physically removed from the sample. The workflow is intuitive: bind the unwanted molecules, pull them away, keep the rest.

Yet capture-based workflows have their own risks. If hybridization is incomplete, some rRNA escapes removal. If the probe set does not cover sequence diversity well, depletion efficiency falls. If capture is too aggressive or specificity is imperfect, non-target molecules can be removed along with the intended ribosomal fraction.

The right depletion logic depends less on vendor chemistry than on how predictable the rRNA background is across the sample. Standard mammalian total RNA with well-characterized targets is very different from bacterial RNA, environmental RNA, or host-pathogen material. In bacterial expression studies, a workflow built around Bacterial RNA Sequencing is often more realistic than forcing a generic depletion design onto diverse inputs. In interaction studies where host and microbial transcripts coexist, Dual RNA-seq becomes the stronger framework because mixed RNA backgrounds are part of the design problem from the start.

What rRNA Removal Actually Changes in the Library

The simplest way to compare workflows is to ask what each one preserves. Poly(A) selection preserves a narrower but cleaner message-centered library. rRNA depletion preserves a broader and often messier transcript universe, but one that may be much closer to the underlying biology.

That difference becomes especially important when the target is not standard mRNA. Many long non-coding RNAs are captured better in depletion-based designs. Histone mRNAs are a classic example of what poly(A)-focused workflows can miss. Replication-dependent histone mRNAs generally lack canonical poly(A) tails, so a poly(A)-selected library can underrepresent them even when they matter biologically. A broader depletion workflow is often better when the goal is to see the real transcript landscape rather than only its polyadenylated subset.

One more point matters here. Incomplete depletion is not merely a cost problem. It is also an interpretation problem. Residual rRNA changes the effective complexity of the library. It can flatten informative coverage, reduce power for low-copy targets, and create uneven comparability between samples if depletion success varies across conditions. In multi-group experiments, inconsistency in depletion can be more damaging than a modest reduction in read depth.

Once depletion succeeds, a second issue appears. Broader transcript retention means the analyst begins to recover more precursor RNA, non-polyadenylated RNA, intron-containing species, and isoform complexity. That is not an optional downstream detail. It is a direct consequence of rRNA handling. The next question is no longer only how much RNA was recovered. It is what kind of transcript architecture has now become visible.

Figure 4. Mechanistic diagram of splicing regulation showing exons, introns, splice sites, branch point, polypyrimidine tract, and the distribution of enhancer and silencer elements such as ESE, ESS, ISE, and ISS.

This is not a detour away from rRNA. It is the downstream consequence of handling rRNA correctly. The moment a library becomes broad enough to retain transcript architecture, analysis has to move beyond simple gene-level counting.

Why Full-Length Resolution Matters After rRNA Background Is Reduced

Once broader transcript retention is achieved, the next limitation is often the read model itself. Short reads are excellent for scale and counting. They are less reliable when the question becomes full-length isoform structure, intron retention, alternative end usage, or hidden exon combinations. In those settings, the problem is not that the RNA is absent. The problem is that its architecture is fragmented across many short observations.

That is why full-length transcript strategies become more attractive after successful rRNA management. If the study needs end-to-end isoform resolution, then long-read or native-molecule designs can reveal what short-read assembly may only approximate. This is particularly important for intron retention, cryptic splicing, non-canonical isoforms, and transcript classes that expand once ribosomal background has been reduced.

For projects where transcript architecture matters more than minimum cost, full-length RNA approaches often become the decisive second layer after broad library preparation. That is where long-read transcript strategies complement rather than replace conventional RNA-seq. The goal is no longer just to count molecules. It is to see which molecules actually exist.

Figure 5. Side-by-side comparison of short-read and full-length transcript sequencing for complex RNA architecture. The figure illustrates how fragmented short reads can leave exon connectivity, intron retention, and hidden isoform combinations ambiguous, while full-length reads resolve transcript structure directly.

The logic is now complete. rRNA handling is not an isolated preprocessing concern. It is the first decision in a chain that determines how much biology becomes visible, which RNA classes enter the library, and whether the resulting complexity can actually be resolved.

rRNA as a Signal in Phylogenomics and Environmental Surveillance

The same molecule that creates problems in transcriptomics remains indispensable in phylogenomics and environmental surveillance. That is not a contradiction. It is a reminder that signal and noise depend on study design.

rRNA genes are uniquely useful because they combine conserved regions with hypervariable regions. Conserved segments provide primer-anchoring sites that work across broad taxonomic groups. Hypervariable segments provide the diversity needed for classification. That dual architecture is why 16S and 18S markers remain central to microbial and eukaryotic community profiling.

For bacteria and archaea, 16S rRNA remains the standard phylogenetic marker. For microbial eukaryotes, 18S rRNA remains widely used. In fungal studies, ITS regions are often added for finer taxonomic discrimination. The underlying logic has remained durable because it is built into the architecture of the molecule itself.

What has changed is the standard of interpretation. It is no longer enough to ask whether a primer pair amplifies. The sharper question is what it systematically misses. Every primer set introduces bias. Some taxa amplify efficiently. Others are underrepresented. Some lineages remain hidden because conserved-region assumptions do not fit their true sequence space. This is one reason microbial dark matter remains a live topic despite years of amplicon sequencing.

That is why full-length marker strategies matter more than before. Short amplicons are efficient and scalable, but they compress positional information. Full-length marker sequencing preserves more sequence context and often improves taxonomic resolution, especially in complex communities or among closely related taxa. When phylogenetic confidence matters, Full-Length 16S/18S/ITS Amplicon Sequencing is often the stronger option. For broader community surveys at scale, 16S/18S/ITS Amplicon Sequencing remains an efficient entry point.

If the question expands beyond who is present to what the community can do, then marker logic is no longer sufficient by itself. That is where Metagenomic Shotgun Sequencing or Long-Read Metagenomic Sequencing becomes the next layer. If the question shifts again from potential to activity, Metatranscriptomic Sequencing becomes the more direct route to active expression.

Environmental surveillance makes this logic even clearer. In wastewater, soil, aquatic systems, and built environments, rRNA-derived loci provide stable anchors for longitudinal comparison. That stability is one reason marker-based profiling remains durable even in the era of long-read metagenomics. It is scalable, standardized, and comparatively easy to benchmark across time points.

Still, marker surveys have limits. They describe composition more easily than activity. That is why many advanced studies now combine methods rather than forcing one assay to answer every question. Amplicon sequencing provides taxonomic scaffolding. Metagenomics expands functional potential. Metatranscriptomics adds active expression. In host-associated settings, Dual RNA-seq can add the interaction layer by capturing both sides of the system.

Conclusion

rRNA is no longer well described as a passive scaffold. It is a catalytic core, a potential regulatory layer, and a dominant technical variable in transcriptomics. Which of those roles matters most depends on the design of the experiment.

If the goal is broad expression profiling, rRNA is often a background burden that should be reduced with care. If the goal is translation biology, ribosome heterogeneity, or modification-aware analysis, rRNA becomes an analyte rather than a nuisance. If the goal is phylogeny or environmental surveillance, rRNA becomes the reference framework that makes taxonomic assignment possible.

The key design mistake is to treat rRNA the same way in every study. The better approach is question-first workflow design: remove rRNA when it blocks sensitivity, preserve transcript diversity when broad architecture matters, and profile rRNA directly when ribosome state is part of the biology.

FAQ

Why is rRNA depletion so important in RNA-seq?

Because rRNA can dominate total RNA input. If it is not reduced, it consumes sequencing reads that should have been used to detect lower-abundance and biologically informative transcripts.

When is poly(A) selection the better choice?

It is often better when the input RNA is high quality and the main goal is standard eukaryotic mRNA expression profiling. It gives a cleaner mRNA-focused library, but narrows transcript diversity.

When is rRNA depletion the better choice?

It is usually better when the study needs broader transcript coverage, including non-polyadenylated RNAs, lncRNAs, histone mRNAs, degraded input, or mixed-background samples.

Can degraded RNA still support effective depletion-based libraries?

Yes, but performance becomes more variable. Degraded RNA changes target accessibility and hybridization behavior, so depletion success depends more heavily on input quality and workflow compatibility.

When should rRNA be profiled directly instead of removed?

It should be profiled directly when the biological question centers on ribosome state, rRNA modification, ribosome heterogeneity, or native translational regulation rather than transcript abundance alone.

Why do RNase H and bead-based depletion behave differently?

Because one removes rRNA by cleavage after RNA:DNA hybrid formation, while the other removes it by physical capture and subtraction. Their efficiencies and failure modes are therefore different.

Can rRNA modifications really affect translation?

Increasing evidence suggests yes, especially for pseudouridylation and 2′-O-methylation. What is better established is the existence of fractional modification and ribosome heterogeneity; transcript-selective consequences remain more system-dependent.

Why are native RNA methods useful for rRNA biology?

Because native RNA workflows preserve molecule-level context that can retain signal linked to RNA chemistry, which is harder to recover from cDNA-based methods alone.

Why do splicing and isoform questions show up in an rRNA article?

Because once rRNA is successfully reduced, broader transcript classes become visible. That means transcript architecture, intron retention, and isoform complexity become downstream interpretation issues.

Why is 16S or 18S rRNA still so widely used?

Because rRNA genes combine conserved and variable regions in a way that makes them powerful markers for taxonomic classification and community profiling.

Can one assay answer transcript abundance, translation state, and rRNA modification all at once?

Usually not well. The strongest studies often combine broad transcript profiling, ribosome-centered assays, and modification-aware methods rather than forcing one platform to answer every layer.

Is rRNA always background in transcriptomics?

No. In conventional RNA-seq it is often background. In ribosome-state studies and phylogenetic studies, it can become central signal.

References:

1. Erales J, Marchand V, Panthu B, et al. Evidence for rRNA 2′-O-methylation plasticity: Control of intrinsic translational capabilities of human ribosomes. Proceedings of the National Academy of Sciences. 2017;114(49):12934-12939. https://doi.org/10.1073/pnas.1707674114
2. Marcel V, Ghayad SE, Belin S, et al. p53 Acts as a safeguard of translational control by regulating fibrillarin and rRNA methylation in cancer. Cancer Cell. 2013;24(3):318-330. https://doi.org/10.1016/j.ccr.2013.08.013
3. Natchiar SK, Myasnikov AG, Kratzat H, Hazemann I, Klaholz BP. Visualization of chemical modifications in the human 80S ribosome structure. Nature. 2017;551:472-477. https://doi.org/10.1038/nature24482
4. de Brouwer JFC, Kroonen JS, Vissers LELM, et al. Variants in PUS7 cause intellectual disability with impaired pseudouridylation and reduced decoding fidelity. American Journal of Human Genetics. 2018;102(3):526-539. https://doi.org/10.1016/j.ajhg.2018.02.004
5. Krogh N, Jansson MD, Häfner SJ, et al. Profiling of 2′-O-Me in human rRNA reveals a subset of fractionally modified positions and provides evidence for ribosome heterogeneity. Nucleic Acids Research. 2016;44(16):7884-7895. https://doi.org/10.1093/nar/gkw482
6. Johnson JS, Spakowicz DJ, Hong BY, et al. Evaluation of 16S rRNA gene sequencing for species and strain-level microbiome analysis. Nature Communications. 2019;10:5029. https://doi.org/10.1038/s41467-019-13036-1
7. Yarza P, Yilmaz P, Pruesse E, et al. Uniting the classification of cultured and uncultured bacteria and archaea using 16S rRNA gene sequences. Nature Reviews Microbiology. 2014;12(9):635-645. https://doi.org/10.1038/nrmicro3330

This resource discusses ribosomal RNA biology and sequencing workflow design for research use contexts only. Method selection should follow the experimental question, sample type, and analytical scope.