Summary

GLORI-seq protocol sample requirements define the feasibility boundary for low-input embryo RNA m6A mapping projects. This guide helps developmental biology researchers evaluate whether their RNA quantity, integrity, and library preparation strategy are sufficient for GLORI-seq before committing scarce embryonic material. Topics covered include input thresholds by protocol version, integrity benchmarks, rRNA depletion versus poly-A enrichment trade-offs, QC risk flags at each workflow stage, and paired study design logic for teams working with limited samples.

Key Takeaways

• GLORI-seq delivers single-nucleotide, quantitative m6A stoichiometry without antibody pull-down, but the chemical conversion step imposes stricter RNA integrity requirements than immunoprecipitation-based methods such as MeRIP-seq
• Updated GLORI 2.0 and 3.0 protocols (Sun et al., Nature Methods, 2025) significantly extend low-input applicability: GLORI 2.0 is compatible with RNA from approximately 10,000 cells; GLORI 3.0 employs a reverse transcription-silent carrier RNA to reach as low as 500–1,000 cells
• Embryo and low-cell-number samples face specific input and integrity challenges that must be assessed before library preparation begins
• rRNA depletion is generally preferred over poly-A enrichment for partially degraded or very-low-input embryo RNA
• Three predictable QC failure points can be identified and mitigated before full cohort commitment
• Paired RNA-seq controls are required for meaningful stoichiometric interpretation; running GLORI-seq alone is not sufficient to report modification rates

Why RNA Input Is the First Question in Any GLORI-Seq Embryo Project

Embryonic samples — preimplantation-stage embryos, oocytes, isolated blastomeres — are among the most constrained inputs in RNA biology research. A single mouse blastocyst yields on the order of nanograms of total RNA. A collection of ten oocytes may not reach the input range that standard RNA-seq protocols accept without amplification.

When the assay in question involves a multi-step chemical conversion, the problem compounds: each processing step that degrades or consumes RNA narrows the feasibility window.

m6A is not a static feature of the embryonic transcriptome. In mice, global RNA m6A levels decrease from the germinal vesicle (GV) oocyte stage to the 2-cell stage, then increase progressively following zygotic genome activation (Zhu et al., Genome Biology, 2023). In zebrafish, over one-third of maternal mRNAs carry m6A marks, and their clearance via the reader protein Ythdf2 is required for timely maternal-to-zygotic transition (Zhao et al., Nature, 2017). These dynamics are not resolvable with peak-level enrichment methods; they require quantitative, site-level data. That is precisely the gap GLORI-seq was designed to fill — and precisely why input constraints matter so much in this context.

Figure 1. End-to-end GLORI-seq workflow from embryo RNA input to site-level m6A stoichiometry. Each step introduces a quality dependency: RNA integrity affects conversion efficiency; conversion completeness determines stoichiometry accuracy; library complexity sets the ceiling on unique-site coverage.

The Chemical Conversion Step: Why It Amplifies Input Sensitivity

GLORI-seq relies on glyoxal- and nitrite-mediated deamination of unmethylated adenosines. Unmodified A residues are converted to inosine, which is read as G during reverse transcription; m6A-modified positions resist conversion and read as A. The stoichiometry of conversion — what fraction of unmodified A residues were successfully converted — directly determines whether a site can be called with confidence.

The original GLORI 1.0 protocol (Liu et al., Nature Biotechnology, 2023) demonstrated transcriptome-wide, single-base quantification at 50 ng mRNA input. However, the deamination conditions were relatively harsh: lengthy reaction times and elevated temperatures that degrade RNA secondary structure. For samples already at the lower bound of detectable quality, this created a compounding degradation problem. A sample that enters the conversion step with a marginal RIN has reduced probability of generating a library with sufficient complexity and coverage to support site-level calling.

How GLORI-Seq Input Demands Compare to MeRIP-Seq

In MeRIP-Seq, the m6A antibody immunoprecipitation step is input-intensive — typically requiring micrograms of total RNA or hundreds of nanograms of polyA-selected RNA — but the underlying RNA is not chemically modified. Fragmentation and immunoprecipitation are gentler on RNA integrity than chemical conversion. Partially degraded RNA can still yield enrichment signals, even if peak calling resolution is reduced.

GLORI-seq inverts this trade-off. Antibody pull-down is eliminated, removing enrichment bias and enabling absolute stoichiometry calculation. But the chemical conversion step introduces a hard dependency on RNA integrity that immunoprecipitation does not. A sample viable for MeRIP-seq peak calling may not be viable for GLORI-seq site-level quantification. This distinction is the single most important factor in pre-submission feasibility assessment for embryo projects.

Feasibility Thresholds: Evaluating Your Embryo RNA Before Submission

Before any embryo RNA sample enters GLORI-seq library preparation, four parameters determine feasibility. These are not arbitrary checkboxes; each maps to a specific failure mode downstream.

RNA Input Range: Defining the Feasibility Window

The development of updated GLORI protocols has substantially lowered the input floor compared to GLORI 1.0.

• GLORI 2.0 (Sun et al., Nature Methods, 2025): Compatible with RNA from approximately 10,000 cells. Operates one to two orders of magnitude below GLORI 1.0 input requirements. Suitable for small dissected tissue regions, sorted cell populations, and pooled early-stage embryos.
• GLORI 3.0 (Sun et al., Nature Methods, 2025): Employs a reverse transcription-silent carrier RNA strategy to achieve high-quality m6A quantification from as few as 500–1,000 cells. This range is relevant for small embryo collections and select preimplantation stages.
• Uli-epic GLORI (an alternative library strategy from independent groups): Demonstrated m6A quantification from 10 ng of rRNA-depleted RNA in sperm and neural stem cell samples. Illustrates that protocol-level engineering, not sample biology alone, determines feasibility.

For teams working below these input ranges, the path forward involves either sample pooling (with the caveat that pooling mixes potentially heterogeneous developmental stages), staged pilot runs to establish library quality before committing a full cohort, or consultation on protocol adaptation.

Risk flag — insufficient input: If total RNA yield from your sample collection falls below the range compatible with the target protocol version, library complexity will be insufficient to support reliable site-level calling. Sequencing more deeply does not compensate for low molecular complexity at the library stage.

RNA Integrity: RIN, DV200, and What They Mean for Embryo Samples

Figure 2. A four-parameter feasibility matrix for pre-submission embryo RNA assessment. Each dimension maps to a specific downstream failure mode: insufficient input limits library complexity; low RIN or DV200 reduces conversion efficiency; chemical contamination inhibits deamination chemistry; multiple freeze-thaw cycles compound integrity loss.

RNA integrity interacts with GLORI-seq feasibility at two points: before conversion (determining whether the input pool contains enough intact transcript for conversion to proceed efficiently) and after conversion (determining whether libraries derived from partially degraded RNA have coverage sufficient for site calling).

The RIN (RNA Integrity Number) remains the most common pre-library integrity metric. For standard bulk RNA-seq, RIN values ≥7 are broadly accepted as suitable. For GLORI-seq, the chemical conversion step shifts this tolerance downward: the amination reaction proceeds less efficiently on already-fragmented RNA, and libraries derived from low-RIN input tend to show higher read duplication rates and lower unique-site coverage.

DV200 — the percentage of RNA fragments longer than 200 nucleotides — provides a more relevant metric for degraded samples, as it captures fragment distribution rather than a single score. For partially degraded embryo RNA where RIN values are uninformative or unavailable (as is often the case with very small input quantities measured on Bioanalyzer High Sensitivity chips), DV200 is a more actionable indicator of library preparation suitability.

Risk flag — low integrity: An RNA sample that scores below the protocol-specific integrity threshold does not simply produce noisier data — it may fail library preparation entirely. Conversion efficiency drops, reverse transcription stalls on short fragments, and PCR amplification over-represents a small subset of molecules. The result is a library that appears complete by concentration metrics but fails depth and diversity filters at the sequencing stage.

RNA Integrity: The Embryo-Specific Problem

Embryo RNA extraction faces challenges absent in standard cell-line or tissue workflows. Total RNA yield from a single preimplantation embryo is typically in the picogram-to-low-nanogram range, requiring extraction protocols (such as SMART-seq2-compatible lysis or single-cell RNA isolation kits) that differ substantially from bulk Trizol or column-based approaches. These protocols introduce different RNA quality profiles. Carrier RNA supplementation during extraction is sometimes used to stabilize yield — but carrier RNA composition must be considered if the same sample will undergo m6A profiling, since exogenous RNA can dilute the signal if not properly controlled.

Risk Flags: When to Pause Before Submitting

The following conditions each represent a predictable failure pathway. Identifying them before submission avoids committing irreplaceable material to a run likely to fail QC:

1. High degradation (DV200 < project threshold or RIN < relevant benchmark): Conversion efficiency drops proportionally with fragment length distribution. Consider re-extraction, alternative stabilization, or protocol-specific adaptation before proceeding.
2. Input below the protocol floor: Running insufficient input does not produce lower-confidence data — it typically produces no interpretable data. Assess whether sample pooling or a different protocol version is feasible.
3. Chemical contaminants from extraction: Trizol carryover, ethanol residues, or guanidinium contamination inhibit the deamination chemistry. A260/A230 ratio below ~1.8 is a meaningful warning sign.
4. Multiple freeze-thaw cycles: Each freeze-thaw cycle degrades RNA integrity incrementally. For embryo samples that have already experienced two or more cycles, pre-submission integrity measurement — not assumption — is required.

If your samples show any of these risk flags, a pre-submission feasibility review can identify whether protocol adaptation, adjusted pooling strategy, or an alternative m6A mapping approach is more appropriate. Explore the full range of available approaches via our RNA modification sequencing services.

Library Preparation Strategy for Low-Input Embryo RNA

Once RNA has cleared the input and integrity assessment, the next decision point is library preparation strategy: poly-A enrichment or ribosomal RNA depletion. For standard, high-input samples from cell lines or abundant tissues, this is often a matter of experimental preference. For low-input embryo RNA, it is a feasibility question.

Figure 3. Library preparation strategy selection for low-input embryo RNA. The primary branch point is RNA integrity: intact poly-A tails and high RIN support poly-A enrichment; degraded, low-input, or non-polyadenylated transcript targets route toward rRNA depletion. In GLORI 3.0-range experiments, carrier RNA strategy interacts with this choice and should be confirmed with the service team.

Poly-A Enrichment: When It Works and When It Doesn't

Poly-A enrichment uses oligo-dT beads to selectively capture polyadenylated mRNAs. The advantages are well-known: high mRNA enrichment efficiency, low rRNA background, and deep per-transcript coverage relative to total RNA approaches.

For GLORI-seq on embryo samples, poly-A enrichment has two meaningful constraints:

• It requires intact poly-A tails. Degraded RNA yields shorter poly-A tails and reduced capture efficiency. For embryo RNA with a RIN below ~7 or a DV200 indicating extensive fragmentation, poly-A capture efficiency drops substantially.
• It excludes non-polyadenylated transcripts. Certain classes of maternal transcripts, regulatory RNAs, and stage-specific isoforms lack conventional poly-A tails. If your research question involves these transcript classes, poly-A enrichment introduces systematic ascertainment bias.

Poly-A enrichment is appropriate for low-input embryo samples when RNA integrity is confirmed to be high (intact poly-A tails), input is sufficient to yield the target mRNA mass after enrichment, and the research question is focused on conventional polyadenylated mRNAs.

rRNA Depletion: The Preferred Route for Degraded or Low-Input Embryo RNA

rRNA depletion (ribosomal RNA removal) retains all RNA species while eliminating the dominant rRNA background. For embryo samples where:

• RNA integrity is moderate or uncertain
• Input is near the low end of the feasibility window
• The research question involves non-polyadenylated transcripts or broad transcriptome coverage

...rRNA depletion is generally the more robust strategy.

The trade-off is that total RNA input requirements for effective rRNA depletion are not negligible — depletion kits have their own input recommendations, and performance degrades with very small input quantities. For extremely low-input embryo samples (GLORI 3.0 range), the carrier RNA approach used in that protocol partially addresses this by supplementing the reaction volume without contributing interpretable m6A signal.

For single-embryo or near-single-embryo applications, the choice between poly-A enrichment and rRNA depletion should be made in consultation with the sequencing service team, as protocol version and carrier RNA strategy interact directly with this decision.

QC Checkpoints and Risk Flags Along the GLORI-Seq Workflow

Three discrete checkpoints govern whether a GLORI-seq run from embryo RNA will produce interpretable data. Failure at any of these does not produce lower-quality results — it produces results that cannot be interpreted. Understanding what each checkpoint measures allows teams to identify potential failures before, not after, sequencing investment.

Pre-Library QC: The Input Gate

Before library construction begins, the following measurements define whether the sample can proceed:

• Quantification: Total RNA mass by fluorometric assay (Qubit or equivalent). Spectrophotometric A260 alone is insufficient for low-input samples due to contaminant interference.
• Integrity profiling: Bioanalyzer or TapeStation trace. RIN, DV200, and fragment size distribution together provide more information than any single metric. For very low-input samples, High Sensitivity chips or Pico chips provide more accurate size distributions.
• Purity: A260/A280 ≥ 1.8–2.0; A260/A230 ≥ 1.8. Values below these ranges indicate protein, phenol, or solvent contamination that will inhibit downstream enzymatic steps.

Risk flag at pre-library gate: Samples that fail purity thresholds should not proceed. Re-precipitation or re-extraction is the appropriate response, not attempting conversion with contaminated input. For low-input samples where re-extraction is not possible, documenting the contamination profile allows the service team to adjust conversion conditions within the range that the chemistry tolerates.

Post-Conversion QC: Detecting Incomplete Amination

After the glyoxal-nitrite deamination step, a critical quality indicator is the A-to-G conversion rate of unmethylated adenosines. In a successful GLORI-seq run, the bulk of unmodified A positions should show conversion to G in the sequencing output. Incomplete conversion — whether due to RNA degradation, suboptimal reaction conditions, or chemical contamination — produces a dataset where m6A and unmodified A are indistinguishable, rendering stoichiometry calculation invalid.

Post-conversion fragment size distribution should also be assessed before library construction proceeds. The conversion step in GLORI 1.0 was noted for causing significant RNA degradation (Liu et al., Nature Biotechnology, 2023). GLORI 2.0 was specifically designed to mitigate this through milder reaction conditions — a key advance for low-input samples where post-conversion fragmentation reduces library complexity below the threshold for reliable calling.

Risk flag at conversion gate: A conversion rate outside the expected range — either too low (incomplete deamination) or too variable across replicates — indicates a technical failure that will propagate into site-calling errors. This checkpoint is why replicate runs and spike-in controls are recommended for first-time embryo sample submissions.

Sequencing-Stage Indicators of a Successful GLORI-Seq Run

At the sequencing output stage, the following metrics distinguish a technically successful run from one that passed library QC by concentration but failed in coverage:

• Unique mapping rate: Low-input libraries from degraded embryo RNA show elevated PCR duplicate rates, which reduce effective coverage. Post-deduplication read counts determine whether sufficient unique molecules were sequenced to support per-site stoichiometry.
• Q30 base quality: Illumina PE150 sequencing is recommended; Q30 ≥ 90% is the benchmark for GLORI-seq data used in the original and updated protocols.
• Site-calling thresholds: m6A sites are called based on A-retention rate at each position relative to background A-retention in unconverted regions. Sites with insufficient coverage depth cannot be called with statistical confidence regardless of sequencing depth; the constraint is unique molecule count at each position.

Figure 4. Three-stage QC checkpoint framework for GLORI-seq runs from low-input embryo RNA. Each gate has distinct pass indicators and failure signatures. Failure at the pre-library gate prevents wasted conversion reagents; failure at the conversion gate signals chemistry issues rather than sequencing depth problems; sequencing-stage failures indicate insufficient library complexity, which cannot be recovered by deeper sequencing.

Study Design for Low-Input Embryo m6A Projects

The decisions made during study design — not just sample preparation — determine whether a GLORI-seq project produces actionable biology or an uninterpretable dataset. For embryo projects where sample scarcity is the defining constraint, study design deserves the same rigor as sample preparation.

Why Paired RNA-Seq Is Not Optional for Stoichiometric Interpretation

GLORI-seq measures the fraction of a given position that carries m6A — the modification stoichiometry. That fraction is calculated as: A-retention rate at the site divided by the background A-retention rate in the same sample.

To convert this to a biologically meaningful value — how much of transcript X is m6A-modified at position Y in embryo stage Z — the total abundance of that transcript must be known. Total transcript abundance is not measured by GLORI-seq alone. It requires a paired RNA-seq library from the same RNA input.

Without this, stoichiometry values are technically derivable but biologically uninterpretable: a site with 30% modification rate on a lowly expressed transcript has a different biological significance than 30% modification on a highly expressed one. In early embryogenesis, where transcript abundance changes by orders of magnitude between developmental stages, this distinction is critical.

Running GLORI-seq without a paired RNA-seq control is a common design error in low-input projects, often driven by the desire to conserve sample. The irony is that it renders the GLORI-seq data uninterpretable, wasting the entire sample allocation.

Replication, Pooling, and Staged Project Design for Scarce Samples

For embryo projects, three design strategies help maximize the information yield from limited material:

Biological replication: A minimum of two to three independent biological replicates is required for any m6A site to be called with statistical confidence across a comparison. Pooled samples from multiple embryos can serve as a single replicate — but the pooling strategy must be consistent across all replicates, and the developmental stage composition of each pool must be verified.

Staged project structure: Rather than committing a full sample cohort to a single run, a staged approach — pilot with a small number of samples to validate library quality and conversion efficiency, then proceed to full cohort — limits irreversible sample loss to pilot quantities. This is particularly important when working with embryo stages that require significant experimental effort to collect.

Alternative methods for orthogonal validation: For sites identified by GLORI-seq as high-priority candidates, targeted validation using ONT Direct RNA Sequencing provides an orthogonal, antibody-independent single-molecule readout. ONT direct RNA sequencing reads RNA modifications at the single-molecule level and can confirm stoichiometry estimates from GLORI-seq without additional sample consumption if RNA is available.

Designing a GLORI-seq project with limited embryo material? Our team can review your sample specifications and recommend a protocol version, pooling strategy, and QC plan before you commit your cohort. Contact us to request a pre-submission feasibility review →

Frequently Asked Questions