tRNA Sequencing Service - Demethylation-Optimized, Quantitative Profiling of Canonical and Modified tRNAs
Transfer RNAs are the most heavily modified and structurally stable RNA species in the cell, averaging roughly 13 chemical modifications per molecule. Standard small-RNA library preparation routinely under-represents heavily modified tRNAs because reverse transcriptase stalls or drops off at modified bases, producing tRNA profiles that look quantitative but are not.
CD Genomics runs a demethylation-optimized tRNA sequencing workflow that combines engineered AlkB-family demethylase treatment with a highly processive reverse transcriptase, removing the major source of RT-blocking modifications before library construction. The result is full-length cDNA coverage across the majority of tRNA isodecoders, accurate relative quantification, and modification-site mapping from the same dataset.
Key Highlights:
Demethylase pretreatment removes m1A, m1G, m3C, and m2,2G — the modifications most responsible for RT drop-off and tRNA under-quantification.
Clustered-reference mapping resolves multi-copy tRNA genes and near-identical isodecoders that standard genome aligners discard as multimappers.
Compatible with low-input and degraded samples, including FFPE-derived RNA, through optimized library chemistry.
Applicable to any species with a curated or predictable tRNA gene set, from bacteria to mammals.
Transfer RNAs (tRNAs) are small, highly structured non-coding RNAs that physically link the genetic code to protein synthesis. Each tRNA folds into a conserved cloverleaf/L-shaped structure, carries a specific amino acid at its 3' CCA end, and reads a codon on the mRNA through its anticodon loop inside the ribosome. tRNA sequencing (tRNA-seq) is the set of next-generation sequencing methods built to profile this transcript population at isodecoder resolution — distinguishing tRNA genes that share an anticodon but differ in body sequence — while also capturing the dense layer of post-transcriptional modifications that standard RNA-seq pipelines are not designed to handle.
The technical difficulty is well documented in the literature: tRNA genes exist in dozens to thousands of near-identical copies per genome, and the transcripts themselves carry, on average, around 13 modified nucleotides each, several of which directly interfere with reverse transcription. A generic small-RNA-seq library prep run on total RNA will sequence preferentially through the least-modified, least-structured tRNAs and systematically miss or under-count the rest — a bias that looks like real biology unless the method specifically corrects for it.
CD Genomics' tRNA sequencing service is built around that correction. For tRNA-derived fragment and stress-induced tiRNA profiling instead of full-length tRNA quantification, see our tRF & tiRNA Sequencing service; for single-molecule modification calling at base resolution, see mim-tRNA-seq.
Full-length tRNA structure, direct modification calling, low-complexity samples
General small-RNA profiling, not tRNA-focused
Targeted validation of 1-2 candidates
Legacy screening, low resolution
Technical Advantages
Demethylase-enhanced library prep
AlkB-family demethylase treatment removes m1A, m1G, m3C, and m2,2G prior to reverse transcription, substantially reducing RT stalling and restoring coverage to the 5' end of heavily modified tRNAs.
Clustered-reference, isodecoder-aware mapping
Reads are mapped against a manually curated, clustered tRNA reference rather than a generic genome index, recovering signal from near-identical tRNA gene copies that standard pipelines discard as multimappers.
Low-input and degraded sample compatibility
Optimized adapter ligation and amplification chemistry support reduced RNA input, with dedicated handling paths for FFPE-derived RNA and other degraded sample types.
Oxford Nanopore long-read option for direct modification calling
For projects requiring native-RNA ionic-current-based modification calling or full-length tRNA reads without RT chemistry, we offer Oxford Nanopore direct RNA and cDNA library paths as a complement to Illumina short-read quantification.
Together, these design choices move tRNA-seq from a qualitative presence/absence readout to a quantitative, isodecoder-resolved measurement that supports defensible differential expression and modification-dynamics conclusions.
tRNA Sequencing Workflow Overview
Our workflow is built specifically around the structural and chemical obstacles that make tRNA difficult to sequence accurately.
Total RNA Extraction and Size Selection – Total RNA is isolated and the tRNA-sized fraction (roughly 60-100 nt) is enriched away from larger RNA species.
Demethylase Treatment – The tRNA fraction is treated with engineered AlkB-family demethylases to remove RT-blocking base methylations prior to library construction.
Adapter Ligation and Reverse Transcription – Pre-adenylated adapters with randomized ends are ligated, followed by reverse transcription with a highly processive, structure-tolerant reverse transcriptase.
Library Amplification and QC – cDNA is amplified, indexed, and quality-checked by Bioanalyzer/TapeStation and qPCR before pooling.
High-Throughput Sequencing – Libraries are sequenced on Illumina NovaSeq platforms as the standard path for high-depth isodecoder quantification. For native modification calling or full-length read resolution, we additionally support Oxford Nanopore direct RNA and cDNA sequencing; both platforms can be run in parallel on the same sample set where both data types are needed.
Bioinformatics Analysis – Reads are mapped to a clustered tRNA reference, quantified per isodecoder, and screened for modification signatures; Nanopore ionic-current signals are processed with platform-appropriate modification-calling tools.
Sample QC Criteria and Acceptance Standards
Because tRNA is structurally compact and resistant to degradation relative to mRNA, conventional RNA integrity metrics alone can be misleading. We apply the following acceptance criteria before committing a sample to library preparation.
Parameter
Standard Sample
Low-Input / FFPE Sample
Total RNA quantity
≥ 2 µg
≥ 200 ng (consultation required)
Concentration
≥ 200 ng/µL
≥ 20 ng/µL
OD260/280
1.8-2.3
1.7-2.3
OD260/230
≥ 2.0
≥ 1.7
RIN (where applicable)
≥ 6.5, 28S:18S ≥ 1.0
Not required — DV200 evaluated instead
Visible degradation
None
Expected; evaluated case by case
Sample Storage: RNA should be dissolved in ethanol or RNase-free ultrapure water and stored at -80°C, avoiding repeated freeze-thaw cycles.
Shipping: RNA samples should be sealed in 1.5 mL Eppendorf tubes with sealing film and shipped on dry ice (5-10 lb per 24 hours of transit).
For samples that fall outside standard criteria — degraded clinical material, low-yield biopsies, or ultralow-input preparations — our team reviews QC data individually before confirming feasibility, rather than rejecting samples by threshold alone.
Bioinformatics Analysis Pipeline
Analysis Type
Content Description
Illumina Short-Read Analysis
1. Raw data QC and adapter trimming
Quality filtering, adapter removal, and molecular barcodes extraction where applicable.
2. Clustered-reference mapping
Reads are aligned to a curated, clustered tRNA gene reference that collapses near-identical isodecoders to representative groups.
3. tRNA gene prediction, curation, and annotation
Genomic tRNA gene loci are predicted, intron-containing genes flagged, and reference sets manually refined.
4. Isodecoder-level quantification
Read counts per cluster are normalized to total mapped reads for relative tRNA abundance.
5. Differential expression analysis
Statistical comparison of isodecoder abundance between experimental groups.
6. Modification-site mapping (Illumina)
Position-wise misincorporation and RT-stop fraction analysis to flag putative modification sites; m5C detection via bisulfite-conversion arm.
7. tRNA mutation/variant/isoform detection
Identification of sequence variants and isoforms relative to the reference tRNA set.
8. Codon usage correlation
Correlation of tRNA abundance with genomic codon usage frequency to evaluate translational supply-demand balance.
High-accuracy basecalling (Dorado) with per-read quality filtering; adapter trimming for cDNA libraries.
10. Long-read tRNA mapping
Full-length read alignment to the clustered tRNA reference, enabling unambiguous isodecoder assignment from single reads rather than short-read inferences.
11. Direct modification calling from ionic-current signal
Nanopore-specific modification detection (e.g., m6A, m1A, pseudouridine) from raw signal using modification-aware basecallers, without requiring chemical pretreatment for those sites.
12. Cross-platform modification concordance
Where both Illumina and Nanopore data are generated from the same samples, modification calls are cross-validated between misincorporation-based and ionic-current-based evidence.
For samples where modification chemistry itself — not just expression — is the primary question, we also offer orthogonal tRNA Modification Analysis by Mass Spectrometry, which can be run alongside sequencing-based modification calls for independent confirmation.
Technical Applications
Accurate, isodecoder-resolved tRNA profiling supports research questions that generic small-RNA-seq cannot reliably answer.
Translational supply-demand and codon usage studies
Correlate tRNA isodecoder abundance with codon usage in expressed transcripts to identify translational bottlenecks relevant to proliferation, stress response, or codon-optimized transgene design.
Stress, disease, and developmental tRNA reprogramming
Track condition-specific shifts in tRNA expression and modification levels across treatment, disease, or developmental time-course designs.
Non-canonical tRNA function and microbial physiology
Profile tRNA pools in non-model organisms and microbial communities, where tRNA gene content is phylogenetically informative and tightly linked to growth and translational physiology. Supports an unrestricted range of species, from environmental microbes to mammalian systems.
Modification mapping for tRNA disease research
Map RT-blocking modification signatures across the tRNA pool to support research into tRNA modification enzyme deficiencies and their downstream effects on translation.
Integration with tRNA-derived fragment and translatome studies
Pair full-length tRNA-seq with tRF & tiRNA Sequencing or Ribo-Seq to connect tRNA pool composition directly to fragment biogenesis and active translation.
Deliverables:
FastQ files and BAM alignment files against the clustered tRNA reference
Coverage summary and per-isodecoder quantification tables
Representative outputs from internal validation of our demethylation-optimized tRNA sequencing workflow.
Figure 1. Coverage improvement after demethylase treatment
Read coverage across a representative tRNA isodecoder cluster is shown for mock-treated and demethylase-treated libraries, illustrating restored 5'-end coverage and reduced RT drop-off after modification removal.
Figure 2. Isodecoder-resolved expression profile
Relative abundance across major tRNA isodecoder clusters, demonstrating quantitative separation between low- and high-abundance isoacceptor families.
Figure 3. Modification-site signature mapping
Position-wise misincorporation rates flag canonical modification sites, consistent with known tRNA modification hotspots at the D-loop and T-loop positions.
Restored Coverage Uniformity
Demethylase pretreatment reduces the systematic under-representation of heavily modified tRNAs seen in standard small-RNA-seq.
Modification signatures are extracted from the same sequencing data used for expression quantification, with no separate assay required.
Cross-Species Applicability
The same workflow scales from microbial tRNA pools to complex mammalian tissue panels.
Case Study
Background
Most tRNA expression studies treat the tRNA pool as static. Rappol et al. set out to determine whether tRNA isodecoder expression and modification actually change across early vertebrate development, using zebrafish as a fast-developing model organism, and developed an optimized sequencing protocol (tRAM-seq) to make that measurement possible at genome scale.
Methods
RNA was collected from zebrafish eggs and embryos spanning the 4-cell stage through 24 hours post-fertilization, plus adult ovary tissue. Each tRNA fraction was split three ways: untreated for modification detection, AlkB-demethylase-treated for accurate quantification, and bisulfite-converted for m5C detection. Libraries were reverse-transcribed with a structure-tolerant reverse transcriptase, sequenced on Illumina, and mapped against a manually curated reference built from roughly 20,000 predicted zebrafish tRNA gene candidates.
Results
After clustering near-identical gene copies by sequence similarity and multimapping behavior, the approximately 20,000 candidate tRNA genes collapsed into 68 distinct expression clusters, including all 22 mitochondrial tRNAs. Demethylase treatment abolished the high RT-error signal previously seen at known modification sites m1G9 and m1A58 in a representative Gln-CTG/TTG cluster, confirming that those signals reflected real modifications rather than mapping artifacts. The bisulfite-converted libraries additionally detected m5C modification at positions 49 and 50, matching the expected vertebrate tRNA modification pattern. Across developmental stages, the team observed a clear, coordinated shift in tRNA isodecoder expression and modification profile occurring around the onset of gastrulation.
Conclusion
This work demonstrates that demethylation-optimized tRNA sequencing can resolve isodecoder-level expression changes and pinpoint specific modification sites within a single dataset, directly supporting the kind of developmental and stress-response tRNA reprogramming studies our service is built for.
Why does tRNA sequencing need a different protocol than standard small RNA-seq?
tRNAs carry an average of about 13 modifications per molecule, several of which block or stall reverse transcriptase. A generic small-RNA-seq protocol will under-sequence the most heavily modified tRNAs, producing an expression profile skewed toward the least-modified species rather than a true quantitative picture.
How does demethylase treatment improve the data?
Engineered AlkB-family demethylases remove specific RT-blocking methylations (m1A, m1G, m3C, m2,2G) from the tRNA fraction before library preparation, allowing reverse transcriptase to read through positions it would otherwise stall at, which restores coverage and improves quantitative accuracy.
Can this service handle low-input or FFPE-derived samples?
Yes. We support reduced-input library chemistry for samples as low as 200 ng of total RNA, and we evaluate FFPE-derived and other degraded samples individually using DV200 rather than RIN, since tRNA's compact structure makes standard integrity metrics less informative for this RNA species.
What species can be analyzed?
The method is species-agnostic and has been applied from environmental microbes through mammalian systems. For non-model organisms without an existing curated tRNA reference, our bioinformatics team builds a clustered reference from genomic tRNA gene predictions as part of the analysis.
Can tRNA-seq detect modification sites, or do I need a separate assay?
Modification sites that interfere with reverse transcription leave a misincorporation or RT-stop signature in the same sequencing data used for expression quantification, so modification mapping is included as part of standard bioinformatics analysis. For absolute modification stoichiometry, we also offer orthogonal mass-spectrometry-based modification profiling.
How is tRNA-seq different from mim-tRNA-seq or tRF/tiRNA sequencing?
Standard tRNA-seq quantifies full-length mature tRNA isodecoders and infers modification sites from sequencing error patterns. mim-tRNA-seq uses a different reverse transcriptase chemistry to call individual modifications with higher specificity, while tRF & tiRNA sequencing targets the shorter tRNA-derived fragments rather than full-length tRNAs; we can advise which method, or combination, fits your specific question.
Figure 1. Coverage improvement after demethylase treatment
Figure 2. Isodecoder-resolved expression profile
Figure 3. Modification-site signature mapping
