Differential chromatin accessibility between pre- and post-natal stages highlights putative causal regulatory variants in pig skeletal muscle

By integrating ATAC-seq, molQTL, and GWAS data across 202 porcine skeletal muscle samples, this study reveals distinct developmental shifts in chromatin accessibility between fetal and postnatal stages, identifying stage-specific regulatory variants that prioritize causal mechanisms for agronomically important traits.

The Gist

Imagine the DNA inside a pig's muscle cells as a massive, ancient library. This library contains every instruction needed to build a pig, from its size to how much muscle it has. But here's the catch: not every book in the library is open and ready to be read at the same time.

This paper is like a team of librarians who went into this library at two very different times in a pig's life: before it was born (the fetal stage) and after it was born (the piglet stage). They wanted to see which "books" (genes) were open for reading and which were closed, and how that changed as the pig grew.

Here is the story of what they found, explained simply:

1. The Two Different "Reading Rooms"

Think of the pig's muscle development as a construction project.

• The Fetal Stage (The Blueprint Phase): When the pig is still a baby in the womb, the library is chaotic and full of activity. The librarians found that the "open books" were mostly on the shelves near the entrance (promoters) and in the wide open spaces between the shelves (intergenic regions).
  • What this means: The baby pig is frantically reading the "how-to" manuals for building the basic structure of the muscle. It's all about setting up the foundation, making new cells, and getting the machinery ready.
• The Piglet Stage (The Finishing Phase): Once the pig is born and starts growing, the library changes. The librarians noticed that the "open books" shifted. Now, the action was mostly inside the books themselves (introns).
  • What this means: The foundation is built. Now, the piglet is fine-tuning the details. It's reading the specific chapters on how to make the muscle fibers strong, how to burn fat for energy, and how to contract (move).

2. The "Switches" That Control the Books

The researchers didn't just look at which books were open; they looked for the light switches that turn the lights on in those rooms. These switches are tiny changes in the DNA code called variants.

They discovered something fascinating:

• The Big Switches: The most powerful switches (the ones that really make a difference) were mostly found in the Fetal "Reading Room."
• The Analogy: Imagine you are building a house. The most critical switches are the ones you install while the walls are being built (fetal stage). Once the house is built (piglet stage), you mostly just adjust the dimmer switches for the lights inside the rooms.
• The Result: This suggests that if you want to change how a pig grows or how much meat it produces, the most important genetic changes happen before the pig is even born.

3. Connecting the Library to the Farmer's Goals

Farmers care about specific traits: How much meat does the pig have? Is it lean? How many piglets are born in a litter?

The researchers took their list of "open books" and "powerful switches" and matched them against a giant database of pig traits (like a "Wanted Poster" for good farming traits).

• They found that while the goals (like having a big, lean pig) are the same for both fetuses and piglets, the workers doing the job are different.
• The Fetal Team: Uses a specific set of genes to build the muscle structure.
• The Piglet Team: Uses a different set of genes to make that muscle work and store energy.
• The Surprise: Only three genes were on both teams! This means the biological "recipe" for a pig's muscle changes completely as it grows up.

4. Why This Matters

Think of this research as finding the master key to the pig's genetic library.

For a long time, scientists thought about pig genetics mostly by looking at adult pigs. This paper says, "Wait a minute! The most important changes happen when the pig is still a baby in the womb."

The Takeaway:
If farmers or scientists want to breed better pigs (more meat, better health), they shouldn't just look at the adult. They need to look at the prenatal stage. By understanding which "switches" are flipped during the fetal stage, they can predict and influence the pig's future traits much more effectively.

In a nutshell: The pig's muscle is built in the womb using one set of rules, and then refined after birth using a completely different set. The most powerful genetic levers are pulled before the pig is even born.

Technical Summary

1. Problem Statement

While the Functional Annotation of ANimal Genomes (FAANG) project has extensively annotated chromatin states in livestock, the regulatory landscape of porcine skeletal muscle across prenatal (fetal) and postnatal (piglet) developmental stages remains incompletely characterized.

• The Gap: Existing studies often focus on single time points or specific breeds, lacking a comprehensive, cross-breed integration of chromatin accessibility dynamics.
• The Challenge: It is unclear how chromatin accessibility shifts between developmental stages and how these shifts relate to molecular quantitative trait loci (molQTLs) and complex agronomic traits (e.g., growth, muscle mass, reproduction). Deciphering which regulatory variants are causal for these traits at specific developmental stages is critical for improving pig breeding programs.

2. Methodology

The study employed a multi-layered integrative genomics approach combining ATAC-seq, molecular QTL mapping, and GWAS colocalization.

• Data Aggregation & Preprocessing:

  • ATAC-seq: Aggregated 202 bulk ATAC-seq libraries from four public datasets (GENE-SWitCH, Salavati et al., Miao et al.) covering diverse breeds (Large White, Landrace, Duroc, etc.) and time points (fetal days 30–90; piglets from birth to 6 months).
  • Processing: Reads were aligned to the Sus scrofa 11.1 reference genome. A consensus peak set was generated using the nf-core/atac-seq pipeline (requiring peaks in ≥2 replicates).
  • Differential Analysis: Peaks were normalized (loess) and analyzed using limma-voom to identify Differentially Accessible Peaks (DAPs) between fetal and piglet stages, controlling for sex and batch effects (FDR < 0.05).

• Integration with molQTLs (PigGTEx):

  • Intersected consensus peaks and DAPs with cis-molQTLs (eQTLs, sQTLs, eeQTLs, enQTLs, lncQTLs) from the PigGTEx resource.
  • Defined "strong-effect" variants as the top 25% of variants by absolute effect size for each QTL class.
  • Performed permutation tests (1,000 iterations) to assess the statistical significance of overlaps between accessible regions and QTLs.

• Functional & Trait Analysis:

  • GO Enrichment: Analyzed Gene Ontology (Biological Process) terms for genes associated with promoter-accessible eQTLs in each stage.
  • Colocalization: Intersected stage-specific promoter-accessible eGenes with gene–trait colocalization results from Xu et al. (2024), which combined TWAS, SMR, and Bayesian colocalization (RCP > 0.5) across 59 pig populations and 232 traits.

3. Key Results

A. Chromatin Landscape Dynamics

• Consensus Peaks: Identified 170,978 consensus peaks. The majority were intronic (80,676) and intergenic (62,225).
• Differential Accessibility: Identified 132,275 significant DAPs (70,756 fetal-high, 61,519 piglet-high).
• Genomic Distribution Shift:
  • Fetal Stage: DAPs were significantly enriched in promoter-TSS and intergenic regions.
  • Piglet Stage: DAPs were significantly enriched in intronic regions.
  • Interpretation: This suggests a developmental shift from "transcriptional priming" (promoter/intergenic activity) in fetuses to "regulatory refinement" (intronic activity) in postnatal muscle.

B. Integration with molQTLs

• Significant Overlap: Overlaps between accessible regions and molQTLs were highly significant (empirical P < 0.001) for all QTL classes.
• Promoter Bias in Fetuses: Promoter-TSS DAPs showed a strong asymmetry. Fetal-accessible promoters overlapped with significantly more eQTLs (1,948 genes) and strong-effect variants compared to piglet-accessible promoters (716 genes).
• Strong-Effect Variants:
  • Strong-effect molQTLs (top 25%) were significantly enriched within DAPs (Odds Ratios 1.1–1.2).
  • Top 1% Strong eQTLs: Of the strongest eQTLs overlapping promoter-TSS DAPs, 14 out of 18 were located in fetal-accessible promoters, indicating that high-impact regulatory variants are predominantly active during prenatal development.

C. Functional Enrichment (GO)

• Fetal eGenes: Enriched for RNA metabolism, chromatin organization, DNA repair, and ribosome biogenesis (reflecting cell proliferation and differentiation).
• Piglet eGenes: Enriched for striated muscle contraction, lipid metabolism, and cellular respiration (reflecting functional maturation and metabolic specialization).

D. Trait Colocalization

• Complex Traits: Identified 107 fetal-stage and 30 piglet-stage promoter-accessible eGenes colocalized with 12 complex traits (growth, muscle structure, reproduction).
• Distinct Genetic Architecture: Only 3 eGenes (KYAT1, AGO2, KIAA1191) were shared between stages.
• Breed Specificity: Cross-breed meta-analyses (M_) captured broad growth/muscle traits at both stages, while within-breed analyses (Duroc, Landrace, Yorkshire) highlighted lineage-specific signals (e.g., Landrace specific for litter size in fetuses).

4. Key Contributions

1. First Comprehensive Fetal-Piglet Atlas: Created a robust, cross-breed map of chromatin accessibility in porcine skeletal muscle, distinguishing prenatal and postnatal regulatory landscapes.
2. Developmental Shift Characterization: Provided evidence that the regulatory logic of muscle development shifts from promoter/intergenic control (fetal) to intragenic/intronic control (postnatal).
3. Prioritization of Causal Variants: Demonstrated that the strongest regulatory variants (top 1% eQTLs) are disproportionately located in fetal-accessible promoters, suggesting that prenatal chromatin dynamics are critical for defining adult muscle phenotypes.
4. Trait-Gene Linking: Successfully linked stage-specific regulatory elements to complex agronomic traits, revealing that while traits are shared, the underlying causal genes and regulatory mechanisms are largely distinct between developmental stages.

5. Significance

• Breeding Applications: The study provides a framework for prioritizing putative causal regulatory variants for genomic selection. It suggests that focusing on prenatal chromatin accessibility is essential for understanding the genetic architecture of adult meat yield and quality.
• Biological Insight: It clarifies the transition from a proliferative, transcriptionally plastic fetal state to a metabolically specialized, contractile postnatal state, mediated by distinct sets of regulatory variants.
• Methodological Framework: The integrative workflow (ATAC-seq + molQTL + GWAS colocalization) serves as a template for dissecting complex traits in other livestock species and developmental contexts.

In conclusion, the paper establishes that prenatal chromatin accessibility is a primary driver of regulatory variation affecting complex pig traits, offering a new target for functional annotation and breeding strategies.