🧬 biology

Genomic inbreeding coefficient predicts extinction risk

This meta-analysis of over 1,000 populations across 454 species demonstrates that the genomic inbreeding coefficient (FROH) is a significant predictor of extinction risk, with higher FROH levels correlating with greater threat status, particularly in species exhibiting slow life history traits.

Original authors: Yangfan Chen, Xi Wang, Sizhe Hu, Hongbo Wang, Juha Merilä

Published 2026-08-25

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Yangfan Chen, Xi Wang, Sizhe Hu, Hongbo Wang, Juha Merilä

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

In the hidden landscape of our DNA, there are long stretches where the genetic code is identical on both chromosomes, like a mirror image of itself. Scientists call these "runs of homozygosity." They are the footprints of inbreeding, left behind when an individual inherits the same genetic segments from both parents. In wild populations, the accumulation of these identical stretches is a warning sign. It suggests that a population has shrunk, that relatives are mating, and that the genetic diversity needed to survive disease or environmental change is fading. For decades, conservationists have worried that this genetic erosion leads to an "extinction vortex," where a population becomes so small and inbred that it cannot recover, spiraling toward disappearance. But measuring this risk across the animal kingdom has been difficult. Different studies use different tools and methods to find these genetic footprints, making it hard to compare a wolf in Europe with a fish in Asia. The big question has remained: can we actually use these genetic footprints to predict which species are in danger of vanishing?

A team of researchers at The University of Hong Kong, led by Yangfan Chen and Juha Merilä, decided to answer this by looking at the big picture. Instead of studying just one animal, they gathered data from more than 1,000 populations across 454 different species, ranging from mammals and birds to fish and reptiles. They collected over 1,400 separate estimates of genomic inbreeding, creating the largest dataset of its kind. Their goal was to cut through the noise of different scientific methods to see if a clear pattern emerged: do species that are listed as threatened by the International Union for Conservation of Nature (IUCN) carry more of these genetic footprints than those that are safe?

The researchers found that the answer is yes, but with a crucial twist. They discovered that the most significant factor determining how many of these genetic footprints a species has is not just recent population decline, but the species' own evolutionary history and lifestyle. When they accounted for the different ways scientists measure these footprints and the family trees of the animals, a clear trend appeared: species facing a higher risk of extinction generally have higher levels of genomic inbreeding. However, this relationship is not a simple alarm bell that rings the same way for every animal. The study revealed that the "baseline" level of inbreeding varies wildly depending on the type of animal.

A major part of the variation the team uncovered comes from how fast or slow a species lives its life. Animals with "slow" life histories—those that live a long time, grow large, mature late, and have few offspring—tend to carry more of these genetic footprints naturally, even without recent threats. This is likely because their long lives and smaller population sizes over deep time create a background level of inbreeding that is part of their biology. In contrast, animals with "fast" life histories, which reproduce quickly and in large numbers, tend to have fewer of these footprints. This distinction is vital because it explains why previous studies sometimes failed to find a link between inbreeding and extinction risk; they were comparing animals with very different biological baselines without adjusting for them.

The researchers also had to untangle the technical mess created by different laboratories. They found that the tools used to scan the DNA and the specific rules set for what counts as a "long" stretch of identical code can change the results significantly. For instance, using a stricter rule for what counts as a long stretch will naturally find fewer of them. Yet, even after correcting for these methodological differences, the biological signal remained strong. The study confirms that while the tools matter, the biology matters more. The species identity, shaped by millions of years of evolution, is the primary driver of how much inbreeding is found in a genome.

This work suggests that conservationists should stop looking for a single, universal number that defines a "dangerous" level of inbreeding for all animals. A high level of inbreeding in a long-lived, slow-reproducing species might be normal for that group, whereas the same level in a fast-reproducing species could be a dire emergency. Instead, the best way to use this genetic information is to compare a threatened population against its own close relatives or its own historical baseline. By understanding that every species has its own unique genetic starting line, scientists can better spot when a population is sliding off the track. The study does not claim to have solved the mystery of extinction, but it provides a much clearer map for navigating the complex relationship between our DNA and our survival, showing that to save a species, we must first understand its unique story.

Technical Summary: Genomic Inbreeding Coefficient Predicts Extinction Risk

Problem Statement
The genomic inbreeding coefficient, defined as the proportion of the genome in runs of homozygosity ( $F_{ROH}$ ), is a standard metric for assessing inbreeding in wild populations. However, the sources of variance in $F_{ROH}$ estimates remain poorly understood. Previous studies have yielded conflicting results regarding whether $F_{ROH}$ can reliably predict extinction risk, with some finding no association between $F_{ROH}$ and IUCN Red List status. Furthermore, it is unclear to what extent observed variation in $F_{ROH}$ arises from methodological heterogeneity (e.g., sequencing depth, ROH calling algorithms, length thresholds) versus biological factors such as species identity, phylogenetic history, and life-history strategies. This uncertainty hinders the use of $F_{ROH}$ as a comparative baseline for conservation assessments.

Methodology
The authors conducted a meta-analytic synthesis of population-level $F_{ROH}$ estimates to address three primary questions: (1) What factors (methodological, species-level, or population-level) explain the greatest variation in $F_{ROH}$ ? (2) Do threatened species exhibit higher $F_{ROH}$ than non-threatened species? (3) Do species with different life-history strategies differ in $F_{ROH}$ ?

Data Compilation: The study aggregated data from 129 primary empirical studies and four previously published methodologically standardized datasets. The final dataset comprised 1,491 records representing 1,088 wild populations across 454 species (14,000+ individuals), spanning mammals, birds, reptiles, fish, and other taxa.
Data Extraction: Information extracted included mean and standard deviation of $F_{ROH}$ , sample size, sequencing depth, ROH caller (e.g., PLINK, BCFtools), and minimum ROH length thresholds.
Covariates: Conservation status was obtained from the IUCN Red List. Life-history traits (longevity, age at maturity, body mass, fecundity) were compiled from databases (e.g., AnAge, Amniote Life History Database) to construct a "slow life index."
Statistical Modeling: The authors employed Bayesian phylogenetic generalized linear mixed models (PGLMM) using the MCMCglmm package.
- Fixed Effects: Technical variables (ROH caller, length threshold, sequencing depth) and life-history traits.
- Random Effects: Population identity, study identity, and a phylogenetically structured species term (modeled via a phylogenetic covariance matrix derived from TimeTree).
- Missing Data: Multiple imputation was used to handle missing standard deviations and sequencing depths.
- Sensitivity Analyses: Models were re-fitted using subsets of the data (e.g., specific taxonomic groups, specific ROH length thresholds of 100 kbp vs. 1 Mbp, and excluding synthesis datasets) to test robustness.

Key Contributions

Variance Partitioning: The study provides a comprehensive decomposition of $F_{ROH}$ variance, demonstrating that phylogenetically structured species identity is the dominant source of variation, outweighing methodological and population-level factors.
Extinction Risk Association: It establishes a robust positive association between $F_{ROH}$ and IUCN extinction risk after controlling for methodological heterogeneity and phylogenetic non-independence.
Life-History Link: It quantifies the relationship between life-history strategies and genomic inbreeding, showing that species with "slow" life histories (longer lifespan, larger body size, later maturity, lower fecundity) tend to have higher baseline $F_{ROH}$ .
Methodological Insight: It clarifies the specific impacts of technical choices, confirming that higher ROH length thresholds and specific callers (e.g., PLINK vs. BCFtools) systematically alter $F_{ROH}$ estimates.

Results

Sources of Variation: In the full dataset, the phylogenetically structured species component explained the largest proportion of variance (posterior mean ~61.8%). Methodological variables (technical covariates) explained only ~7.7% of the variance. Population and study identity explained 5.8% and 16.7%, respectively.
Extinction Risk: There is a consistent positive association between extinction risk and $F_{ROH}$ . When IUCN status was treated as an ordinal variable (Least Concern to Critically Endangered), each category increase in risk was associated with an increase of 0.16 logit units in $F_{ROH}$ . Threatened species (Vulnerable, Endangered, Critically Endangered) generally exhibited higher $F_{ROH}$ than Least Concern species.
Life-History Traits: A "slow life index" was positively associated with $F_{ROH}$ . Specifically, body mass showed a positive association, and fecundity showed a negative association with $F_{ROH}$ . This pattern held across mammals and birds, though results for fish were inconclusive due to limited sample size.
Phylogenetic Signal: The phylogenetic signal in $F_{ROH}$ was estimated at 0.62 (median 0.72), indicating that closely related species share similar baseline levels of genomic inbreeding.
Methodological Effects: The minimum ROH length threshold had a strong effect; a tenfold increase in the threshold reduced estimated $F_{ROH}$ by approximately 0.82 logit units. PLINK produced higher estimates than BCFtools.

Significance and Claims
The paper claims that $F_{ROH}$ is a valid comparative indicator of extinction risk, but its utility depends on recognizing that absolute values are not universal thresholds. Instead, $F_{ROH}$ must be interpreted relative to a lineage-specific baseline shaped by evolutionary history and life-history strategy.

The authors propose that previous studies failed to detect an association between $F_{ROH}$ and extinction risk due to two factors: (1) a temporal mismatch where recent demographic declines have not yet fully manifested as genomic inbreeding (the "drift-debt" hypothesis), particularly in long-lived species; and (2) a phylogenetic mismatch where high $F_{ROH}$ in certain lineages reflects ancestral demographic history or conserved life-history traits rather than recent anthropogenic pressure.

Consequently, the study argues that conservation assessments should not rely on a single high $F_{ROH}$ value as a diagnostic for extinction risk. Instead, assessments should sample multiple populations, use closely related species to establish comparative baselines, and account for methodological variation and phylogenetic context. This framework aims to reconcile inconsistencies in prior literature and improve the application of genomic inbreeding estimates in biodiversity conservation.

Technical Summary: Genomic Inbreeding Coefficient Predicts Extinction Risk

More like this