The effect of age and sex on the rate of de novo mutations in barn owls

This study analyzes 33 barn owl parent-offspring trios to establish a mutation rate of 5.6 x 10⁻⁹, revealing that fathers transmit twice as many mutations as mothers and that paternal age significantly increases mutation rates, providing direct evidence of germline senescence in birds.

The Gist

Imagine the genome of a living creature as a massive, intricate instruction manual written in a code of four letters (A, C, G, T). Every time a cell divides to create a new one, a copy of this manual is made. Usually, the copying machine is incredibly precise, but occasionally, it makes a typo. These typos are called mutations.

Most of the time, these typos happen in the body's regular cells and die with the individual. But sometimes, a typo happens in the "master copies" used to make babies (the sperm or eggs). These are called de novo mutations, and they get passed down to the next generation, becoming part of the family's new instruction manual.

This paper is like a detective story where scientists investigated how often these typos happen in Barn Owls, and what factors influence them. Here is the breakdown of their findings using simple analogies:

1. The "Typo Rate" (Mutation Rate)

The scientists wanted to know: How many typos does a Barn Owl baby get from its parents?
They looked at 33 families (parents and their babies) and sequenced their DNA.

• The Finding: On average, a baby owl gets about 9 new typos compared to its parents.
• The Analogy: Imagine the owl's genome is a 3-billion-word encyclopedia. Every time a baby is born, the printing press accidentally adds about 9 new, random words that weren't in the parents' books.
• The Comparison: This rate is very similar to other birds that live for a similar amount of time (about 2-3 years). It's like saying, "If you live a short life and reproduce quickly, your 'typo rate' per generation is standard for your group."

2. The "Dad vs. Mom" Bias

One of the biggest questions was: Do moms and dads contribute the same number of typos?

• The Finding: Dads contribute about twice as many typos as moms.
• The Analogy: Think of the mother's eggs as a set of frozen seeds. They are made before the bird is even born and just sit there waiting. They don't get copied often, so they don't accumulate many typos.
  • Think of the father's sperm as a factory assembly line that runs continuously from puberty until death. Every time the factory makes a new product (sperm), it has to copy the manual. The more times the factory runs, the more chances there are for a typo. Since the factory runs constantly, the dad's "manual" accumulates more errors over time.
• The Result: In this study, for every 3 typos a baby gets, 2 came from dad and 1 came from mom.

3. The "Age Factor" (Germline Senescence)

The researchers also asked: Does getting older make you pass on more typos?

• The Finding: Yes, but mostly for dads.
• The Analogy: Imagine the dad's sperm factory again. If the dad is 2 years old, the factory has been running for 2 years. If he is 6 years old, it has been running for 6 years. The longer the machine runs, the more wear and tear (typos) it accumulates.
  • The study found that for every year a father gets older, he passes on about 0.6 extra typos to his offspring.
  • For mothers, the effect was much smaller and not statistically significant in this study. Because their "seeds" are frozen, their age doesn't seem to create as many new typos, though they might accumulate damage from other sources (like radiation or chemicals) over time.

4. The "Type of Typos" (Mutation Spectrum)

Not all typos are the same. Some change a letter to a similar one (like C to T), while others change it to something totally different.

• The Finding: The types of typos dads pass on are slightly different from the types moms pass on.
• The Analogy:
  • Moms seem to pass on more "C to T" typos, especially in specific contexts (like CpG sites). It's as if the frozen seeds have a specific weakness where the ink fades in a certain way over time.
  • Dads pass on more "T to C" typos. This is likely because their factory line is constantly churning out new copies, and the copying machine has a specific habit of making this particular mistake.

Why Does This Matter?

This study is important for a few reasons:

1. Evolution: Mutations are the raw material for evolution. Without them, species can't adapt. Knowing the "typo rate" helps scientists understand how fast birds can evolve.
2. Health: Just like in humans, too many typos can lead to genetic diseases. Understanding that older fathers pass on more mutations helps us understand the risks of having children later in life.
3. Bird Biology: It confirms that birds, like mammals, have a "paternal age effect." Even though birds are very different from humans, the rules of how their DNA copies itself are surprisingly similar.

In a nutshell:
This paper tells us that Barn Owls are like any other animal when it comes to DNA copying errors: they get about 9 new typos per generation. However, the dad is the main source of these errors, and the older the dad is, the more typos he passes on. The mom's contribution is steadier and less affected by her age. It's a reminder that in the game of life, the "factory" that never stops (the dad's sperm production) eventually makes more mistakes than the "frozen archive" (the mom's eggs).

Technical Summary

1. Problem Statement

De novo mutations (DNMs) are the ultimate source of genetic variation and a fundamental driver of evolution. While mutation rates are known to vary across species due to life-history traits (e.g., generation time, sperm competition), there is a lack of robust, direct estimates for most non-mammalian vertebrates.

• Knowledge Gaps:
  • Most existing data comes from a few mammal species (often primates) or captive individuals, limiting generalizability.
  • The influence of parental age on germline mutation rates in birds is unclear, despite evidence of "germline senescence" in mammals.
  • Sex-specific differences in mutation spectra (the types of mutations) and the magnitude of paternal bias in birds remain under-explored compared to mammals.
  • Previous avian studies often relied on single trios or small datasets, making it difficult to distinguish between stochastic noise and biological signals regarding age and sex effects.

2. Methodology

The study utilized a high-throughput sequencing approach on a wild population of Western barn owls (Tyto alba) to directly estimate mutation rates.

• Study System & Sampling:

  • Subjects: 57 individuals (31 males, 26 females) comprising 33 parent-offspring trios from a monitored wild population in Western Switzerland.
  • Pedigree: Verified by field observations and genomic kinship.
  • Sequencing: Whole-genome sequencing (WGS) at high depth (average 54x, range 25–93x). Libraries were prepared using Nextera and PlexWell protocols and sequenced on Illumina NovaSeq 6000 and HiSeq 4000 platforms.

• Bioinformatics Pipeline:

  • Alignment & Processing: Reads were aligned to the barn owl reference genome (v4) using BWA-MEM. Duplicate marking and Base Quality Score Recalibration (BQSR) were performed using GATK best practices.
  • Variant Calling: GATK HaplotypeCaller and GenotypeGVCFs were used to call variants across the autosomal genome (39 linkage groups) and the Z chromosome.
  • Filtering Strategy: A rigorous multi-step filtering process was applied to identify putative DNMs:
    1. Mappability/Repeat Masking: Excluded regions with poor mappability (<95% unique mapping), annotated repeats, and indel-proximal sites.
    2. Quality Filters: Genotype Quality (GQ) and Reference Genotype Quality (RGQ) thresholds set to >60.
    3. Depth Filters: Removed sites with read depths outside the range of $\mu \pm 2\sigma$ (or absolute bounds of 15–2x mean depth).
    4. Technical Filters: Applied stringent GATK metrics (QD, MQ, FS, SOR, etc.).
    5. Mendelian Incompatibility (MI) Selection: Focused on sites where parents were homozygous (either reference or alternative) and the offspring was heterozygous.
    6. Allelic Balance (AB): Filtered for AB between 0.3 and 0.7 in the offspring.
  • Manual Curation: 580 putative mutations were manually inspected using Integrated Genome Viewer (IGV) to remove false positives caused by mis-mapping or mosaicism.
  • Phasing: DNMs were phased to determine parental origin using read-based phasing (WhatHap) and, for 9 individuals, transmission to a third generation.

• Statistical Analysis:

  • Mutation Rate Calculation: $\mu = N_m / [(1-\beta) \times CG \times 2]$, where $N_m$ is the number of true mutations, $\beta$ is the False Discovery Rate (FDR), and $CG$ is the callable genome size.
  • Age Effects: Generalized Linear Mixed Models (GLMMs) using glmmTMB were used to regress the number of phased mutations against parental age, treating trio size as a fixed effect and parent ID as a random effect.
  • Spectrum Analysis: Fisher's exact tests compared mutation spectra between sexes.

3. Key Contributions

• Dataset Scale: One of the largest trio-based mutation rate studies in a non-human vertebrate (33 trios, 57 individuals).
• Wild Population Focus: Unlike many previous studies using captive animals, this study utilizes a wild population with known ages, providing more ecologically relevant estimates.
• Sex-Specific Spectra: Detailed characterization of how mutation types differ between male and female germlines in birds.
• Germline Senescence in Birds: Provided direct empirical evidence that paternal age significantly drives mutation accumulation in birds, a phenomenon previously well-documented in mammals but less clear in avian species.

4. Key Results

• Mutation Rate:
  • The per-generation mutation rate was estimated at $5.6 \times 10^{-9}$ mutations per site.
  • This aligns with other passerine species with similar generation times (2–3 years), such as the zebra finch and great reed warbler.
  • The yearly mutation rate is approximately $2.09 \times 10^{-9}$.
• Paternal Bias:
  • Fathers transmit approximately 2.3 times more mutations than mothers (Male:Female ratio $\alpha \approx 2.3$).
  • This is a lower bias than typically seen in mammals but consistent with avian patterns.
• Mutation Spectra:
  • Overall: Transitions (Ti) dominate over transversions (Tv) with a Ti/Tv ratio of 2.5. The most common mutation is C>T (51%), particularly in CpG contexts.
  • Sex Differences:
    • Females: Showed a higher proportion of C>T transitions, especially in CpG contexts (Ti/Tv ratio = 4.7).
    • Males: Showed a higher proportion of T>C transitions (Ti/Tv ratio = 2.28).
    • The overall spectra were significantly different between sexes ($p = 0.026$).
• Age Effects:
  • Paternal Age: A significant positive correlation exists between paternal age and the number of transmitted mutations. Each additional year of paternal age adds ~0.6 new mutations ($p = 0.006$).
  • Maternal Age: The effect of maternal age was positive but not statistically significant ($p = 0.13$), likely due to a smaller sample size of older mothers and a dataset skewed toward younger females.
• Chromosomal Distribution: Mutations were distributed across all autosomes proportional to their physical length. The Z chromosome mutation rate was consistent with autosomal estimates.

5. Significance

• Evolutionary Biology: Confirms that mutation rates in birds are strongly linked to generation time, supporting the "generation time effect" hypothesis.
• Germline Biology: Provides the first direct evidence of germline senescence (age-related mutation accumulation) in a wild bird species, suggesting that continuous spermatogonial division in birds leads to error accumulation similar to mammals.
• Conservation & Genetics: Highlights that failing to account for parental age in trio-based studies can introduce substantial variance in mutation rate estimates. This is crucial for interpreting genetic load and fitness in wild populations.
• Mechanistic Insight: The observed sex-specific differences in mutation spectra (e.g., higher C>T in females) suggest distinct underlying mechanisms, potentially related to DNA methylation dynamics during oogenesis or differences in DNA repair efficiency, which may be conserved across vertebrates.

In conclusion, this study establishes a robust baseline for avian mutation rates, demonstrating that while the overall rate is conserved across species with similar life histories, the source and accumulation of mutations are heavily influenced by parental sex and age.