Periodic Reporting for period 1 - GINS (Genomics of social structure and its implications for conservation)
Periodo di rendicontazione: 2022-10-01 al 2025-03-31
Around 45,300 species (27.8% of those assessed) are currently threatened with extinction. These include many species that underpin critical ecosystem services, essential to a sustainable global economy and human well-being. Reversing biodiversity loss requires integrating knowledge from different disciplines (e.g. socioecology and evolutionary biology) in order to design efficient conservation strategies. Genomics approaches provide a framework to quantify the health status of species using genetic diversity, inbreeding, and demographic history. However, such approaches have been widely ignoring a defining aspect of many threatened species - social groups - the fundamental units in which many species are organized and reproduce. While certain mating strategies can enhance diversity, inbreeding and reproductive skew in some species can significantly reduce it. The influence of social structure on genomic diversity remains poorly understood and is primarily inferred from long-term studies of a few iconic species. The main purposes of this MSCA-IF action (GINS) were: i) to build a method to simulate genomics data in species subdivided in social groups, ii) to infer socio ecological parameters (e.g. group sizes) directly from genomics data and, iii) raise awareness of how human activity reshapes biodiversity, to shift the way people understand and relate to the environment.
During the reporting period, the project focused on executing an inference model aimed at advancing our understanding of the social structure of species and its impacts on genetic diversity. This resulted in several notable achievements, which I develop below.
The GINS project focused on quantifying inbreeding through runs of homozygosity (ROH) and identity-by-descent (IBD), and inferring parameters like mating systems, dispersal distances, and social group structures. Within this framework, I developed a novel model that: (i) incorporates the complexities of social structure by explicitly incorporating variation in group composition - including differing group sizes and diverse mating configurations (e.g. varying numbers of males and females); and (ii) introduces innovative statistical inference methods to reconstruct social organization and dispersal patterns entirely from genomic data. The main achievements were the following:
- Core simulation tool that incorporates genomic data and validation. This allows to retrieve gene genealogies under different mating systems (WP1)
- Gene trees successfully generated for monogamy, polygyny, and polygynandry (WP1)
- ABC framework developed to infer social structure parameters from genomic data (WP2)
- Inbreeding metrics (ROH, IBD, relatedness) characterized under different mating systems (WP2)
- Software tool completed and being prepared for public release (WP2)
- Developed ABC model for estimating socio ecological parameters (e.g. mating systems, dispersal)
- Lemur research groundwork maintained, collaborations strengthened with future application planned
In short, the results from the GINS project highlight how mating systems and group composition shape the distribution of IBD tracts and ROHs, providing a foundation for inferring ecological parameters such as dispersal and social organization from genomic data.
The GINS project focused on quantifying inbreeding through runs of homozygosity (ROH) and identity-by-descent (IBD), and inferring parameters like mating systems, dispersal distances, and social group structures. Within this framework, I developed a novel model that: (i) incorporates the complexities of social structure by explicitly incorporating variation in group composition - including differing group sizes and diverse mating configurations (e.g. varying numbers of males and females); and (ii) introduces innovative statistical inference methods to reconstruct social organization and dispersal patterns entirely from genomic data. The main achievements were the following:
- Core simulation tool that incorporates genomic data and validation. This allows to retrieve gene genealogies under different mating systems (WP1)
- Gene trees successfully generated for monogamy, polygyny, and polygynandry (WP1)
- ABC framework developed to infer social structure parameters from genomic data (WP2)
- Inbreeding metrics (ROH, IBD, relatedness) characterized under different mating systems (WP2)
- Software tool completed and being prepared for public release (WP2)
- Developed ABC model for estimating socio ecological parameters (e.g. mating systems, dispersal)
- Lemur research groundwork maintained, collaborations strengthened with future application planned
In short, the results from the GINS project highlight how mating systems and group composition shape the distribution of IBD tracts and ROHs, providing a foundation for inferring ecological parameters such as dispersal and social organization from genomic data.
Social structure has been widely recognized to have significant effects on genetic diversity. But the type of social structure (e.g. polygyny, monogyny, cooperative breeding) and dispersal patterns can either enhance or reduce genetic diversity. Therefore, large uncertainties still prevail about how social structure affects genetic diversity, and how species may respond to disruptions of their (social) structure is unknown. The GINS project added crucial results to this gap of knowledge. It advanced our understanding of social structure by contributing the following results:
i) a set of scripts developed in the widely used SLiM simulation language, designed to model genomic data in populations subdivided into social groups under various mating systems. These scripts are adaptable and can be readily applied by other researchers to study genetic dynamics in their own species of interest;
ii) a theoretical framework that predicts the expected distributions of Identity-by-Descent (IBD) tracts and Runs of Homozygosity (ROH) under varying social structures. These expectations serve as a critical baseline for interpreting patterns observed in empirical genomic data.
iii) Further, this work found that the first entry of the Site Frequency Spectrum (SFS) - singleton class - can be used to distinguish between certain (‘extreme’) mating systems, such as monogamy and polygyny. Since this class captures the number of segregating sites where the derived allele is present in only one individual, it is particularly sensitive to recent evolutionary events. However, it is also highly error-prone, especially in the presence of sequencing artifacts or low coverage data. As such, while these initial findings are promising, further investigation is required to fully understand the robustness and applicability of this signal. Ongoing work is focused on exploring additional summary statistics that may complement or enhance the inference of mating and dispersal systems.
iv) Findings from this project underscore the critical role of social structure in shaping genetic diversity and influencing demographic inferences. Specifically, I demonstrated that social organization can bias genetic signals, producing false indications of population expansion - even when such growth has not occurred. These biases arise because social structure affects patterns of gene flow and relatedness, which in turn distort coalescent-based demographic inferences. This result has significant implications for conservation genetics, particularly in endangered species, where misinterpreted signals of population expansion may lead to underestimating extinction risk and misallocating conservation resources.
i) a set of scripts developed in the widely used SLiM simulation language, designed to model genomic data in populations subdivided into social groups under various mating systems. These scripts are adaptable and can be readily applied by other researchers to study genetic dynamics in their own species of interest;
ii) a theoretical framework that predicts the expected distributions of Identity-by-Descent (IBD) tracts and Runs of Homozygosity (ROH) under varying social structures. These expectations serve as a critical baseline for interpreting patterns observed in empirical genomic data.
iii) Further, this work found that the first entry of the Site Frequency Spectrum (SFS) - singleton class - can be used to distinguish between certain (‘extreme’) mating systems, such as monogamy and polygyny. Since this class captures the number of segregating sites where the derived allele is present in only one individual, it is particularly sensitive to recent evolutionary events. However, it is also highly error-prone, especially in the presence of sequencing artifacts or low coverage data. As such, while these initial findings are promising, further investigation is required to fully understand the robustness and applicability of this signal. Ongoing work is focused on exploring additional summary statistics that may complement or enhance the inference of mating and dispersal systems.
iv) Findings from this project underscore the critical role of social structure in shaping genetic diversity and influencing demographic inferences. Specifically, I demonstrated that social organization can bias genetic signals, producing false indications of population expansion - even when such growth has not occurred. These biases arise because social structure affects patterns of gene flow and relatedness, which in turn distort coalescent-based demographic inferences. This result has significant implications for conservation genetics, particularly in endangered species, where misinterpreted signals of population expansion may lead to underestimating extinction risk and misallocating conservation resources.