Data from: A Comprehensive Account of the Breeding Systems of the Fire Ant Solenopsis invicta

Sierra Hale, Walker1; Lacy, Kip2; Ross, Kenneth1; Zeng, Haolin 1

Published Apr 22, 2025; Updated Jul 11, 2025 on Dryad. https://doi.org/10.5061/dryad.vhhmgqp5g

Data files

Apr 22, 2025 version files 2.35 MB

breeding-system-fireant-CSV.zip

55.41 KB
breeding-system-fireant-EXCEL.zip

2.26 MB
README.md

21.46 KB
remating_analysis.R

12.42 KB

Jul 11, 2025 version files 2.35 MB

breeding-system-fireant-CSV.zip

55.41 KB
breeding-system-fireant-EXCEL.zip

2.26 MB
README.md

21.42 KB
remating_analysis.R

12.42 KB

Abstract

When animals reproduce in social groups, the potential for conflict and cooperation is shaped by the number of reproductive individuals (breeders), their relatedness to one another, and division of reproduction among them. These features comprise species’ “breeding systems.” Despite their importance, breeding systems are poorly characterized in most social animals, and detailed accounts for single species are rare. Here, we fully characterize the breeding systems in invasive populations of the fire ant Solenopsis invicta, a species in which a large genetic element (supergene) determines whether a colony has a single queen (monogyne social form) or multiple queens (polygyne form). Colonies of the monogyne form are simple families, and the breeding system is correspondingly straightforward. The breeding system of the polygyne form is complex, with many features still uncharacterized. We conducted a large longitudinal experiment tracking parentage, relatedness, and supergene genotype in semi-natural polygyne colonies. Along with reanalyzed data from previous studies, we show that colony queen number is highly variable, queens generally mate once, nestmate breeders (queens and their mates) are unrelated, and reproductive skew is pervasive, especially for sexual daughters. Uncommon instances of polyandry occur when a queen remates after initially mating with a male bearing the Sb supergene haplotype (associated with low sperm counts). Paternity skew is pronounced and stable, with Sb sperm contributing to a minority of offspring (particularly sexuals). Thus, the supergene not only determines colony queen number, it broadly affects the breeding system, with impacts on colony kin structure and opportunities for reproductive competition.

Dataset DOI: 10.5061/dryad.vhhmgqp5g

Description of the data and file structure

This repository contains datasets associated with the manuscript:

Hale Walker S, Lacy KD, Ross KG, Zeng H. (2025). A Comprehensive Account of the Breeding Systems of the Fire Ant Solenopsis invicta. Ecology and Evolution.

The study investigates the breeding systems of Solenopsis invicta (fire ants), focusing on the polygyne social form. The data include information on queen longevity, reproductive output, parentage, and genetic structure, providing a detailed characterization of reproductive skew and supergene effects in multiple-queen colonies.

File Versions

Each raw data file has an Excel version containing annotations, colors, and notes to aid understanding.

For analytical purposes, most files are also provided in CSV format, which is cleaner and more suitable for statistical analysis.

All files are packaged into two ZIP archives:

breeding-system-fireant-CSV.zip: Contains selected data files in CSV format for easy import into analysis software.
breeding-system-fireant-EXCEL.zip: Contains all raw data files in Excel format with enhanced readability and annotations.

The content in the CSV and Excel versions is identical, except that the Excel versions include additional formatting and explanatory notes.

LIST of DATA FILES

P queen longevity

Data for survivorship curve for polygyne queens in the GA-2023 dataset.

Column Name	Description
`Time (weeks)`	Time point in weeks since the beginning of the observation.
`Queens Alive`	Number of surviving queens at each time point.
`Survival`	Proportion of queens surviving.

Relative wt & fecund of polyandrous vs monandrous queens

Weight and fecundity (oviposition rate) of nestmate monandrous and polyandrous queens. Oviposition rate is the number of eggs laid in a 4-hour oviposition test. Data are from the first four (primary) sampling points; weights were recorded as well at midpoints between the primary points. An "x" indicates that the queen had died. The supergene haplotypes of the male(s) with which each queen mated are shown in parentheses in the last column. Data are from the GA-2023 dataset.

Column Name	Description
`Colony`	ID of the experimental colony.
`Queen ID`	Unique identifier for each queen.
`Sampling point`	Sampling time point (1–4).
`queen_weight_mg`	Queen weight in milligrams at the given sampling point.
`fecundity`	Number of eggs laid in a 4-hour oviposition assay.
`Mating_status_total`	Full description of mating status, including number and genotype of mates.
`male_supergene_haplotypes`	Supergene haplotypes of all males the queen mated with.
`mating_status`	Mating status: either `monandrous` or `polyandrous`.

Relative pupa production of polyandrous vs monandrous queens

Relative worker and gyne offspring production of nestmate monandrous and polyandrous queens. Values are proportions of daughters of each caste attributable to each nestmate queen at each of the first four (primary) sampling points. An "x" indicates that the queen had died. The supergene haplotypes of the male(s) with which each queen mated are shown in parentheses in the last column. Data are from the GA-2023 dataset.

Column Name	Description
`Colony`	ID of the experimental colony.
`Queen ID`	Unique identifier for each queen.
`Sampling point`	Sampling time point (e.g., 1–4).
`worker pupal offspring`	Proportion of total worker pupae attributed to the queen at the sampling point.
`gyne pupal offspring`	Proportion of total gyne pupae attributed to the queen at the sampling point.
`Mating status (male supergene haplotypes)`	Queen mating status with number and type of male haplotypes (e.g., monandrous/polyandrous with SB/Sb).

Ref pop and parental gtypes-msats NE GA

Microsatellite genotypes and haplotypes for reproductive queens and males used as the reference population for estimating genetic relatedness between and within groups of individuals in the GA-2018 and GA-2023 datasets featured in Hale Walker et al. (2025). The GA-2018 dataset was developed for the study of Ross & Shoemaker (2018), while the GA-2023 dataset was developed for the Hale Walker et al. (2025) study. The genotypes/haplotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele. These are also the parental genotypes/haplotypes for the progenies studied in Hale Walker et al. (2025). Two of the loci (C294, i-126) are located on Chr16 and are in strong LD with the social supergene, which is on the same chromosome; the remaining 11 loci are located on other chromosomes.

Column Name	Description
`Source study`	Publication or study where the sample originated (e.g., Ross & Shoemaker 2018).
`Sample type`	Type of sample (e.g., male mate, reproductive queen).
`Source colony`	Colony from which the individual was sampled.
`Individual`	Unique ID for each individual.
`Sol_49` to `i-126`	Microsatellite genotypes at each locus, in Genepop 3-digit format.
`notes`	Comments on specific genotypes (e.g., null allele flagged).

Raw genotype data for all embryo progeny_GA-2018 dataset (only available in xlsx)

Genotypes of embryos at 14 microsatellite markers and at Gp-9 in polygyne Solenopsis invicta progenies, and of the mother queens and their mates (from Ross & Shoemaker, 2018, BMC Genetics). With the exception of Gp-9, the data are in GENEPOP format with 3-digit alleles (missing alleles are indicated by "000"). Gp-9 genotypes comprise the two alleles (B, b). The loci "red ant" and "C294" each were scored in only a portion of the progenies (cells with dash = unscored; red ant is not polymorphic enough and C294 had not been discovered). Embryos are assigned to patrilines in polyandrous queens (color-shaded rows help to visualize distinct patrilines).

Column Name	Description
`Embryo code`	Unique identifier for each embryo (e.g., 14-1-a,1)
`red_ant` to `i_120`	Microsatellite genotype data at each locus in 3-digit Genepop format
`Gp-9`	Genotype at the Gp-9 locus (e.g., BB, Bb, or missing)
`embryo viability`	Developmental outcome of each embryo (e.g., viable or not viable)

Raw genotype data for all pupa progeny_GA-2023 dataset (only available in xlsx)

Progeny microsatellite genotypes for female pupae from the GA-2023 dataset. The female castes (workers, gynes) are shown separatly and the matrilines (and patrilines for polyandrous queens) to which each sample was assigned are indicated. The genotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele.

Two of the loci (C294, i-126) are located on Chr16 and are in strong LD with the social supergene, which is on the same chromosome; the remaining 11 loci are located on other chromosomes.

Entries lacking a Patriline ID indicate that the corresponding queen was singly mated, and thus no separate Patriline ID was assigned.

Column Name	Description
`Patriline ID`	ID of the patriline (paternal family) to which the individual was assigned.
`Colony ID`	Identifier for the experimental colony.
`Mother (matriline) ID`	Identifier for the maternal queen of the sampled daughter.
`Indiv. daughter ID`	Unique ID for the sampled daughter (e.g., P18-01-S1-W-14).
`C294` to `i_120`	Microsatellite genotype data in Genepop 3-digit allele format.
`Daughter caste`	Caste of the individual (`worker` or `gyne`).
`Sampling point`	Sampling time point (e.g., 1 = first sampling).

worker & gyne zyme genotypes 31 Walton P cols

Allozyme genotypes for workers and gynes compiled from a study of a polygyne population in northern Georgia (Walton Co.) conducted by Ross (1993). Worker genotypes were used to estimate FIS, while data for each caste were used separately to estimate average pairwise nestmate relatedness using all life stages combined for each caste (Hale Walker et al., 2025). The genotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele. Note that different sets of loci were scorable in the different castes/life stages.

Column Name	Description
`Colony`	Identifier for the colony from which the individual was sampled.
`caste`	Caste of the individual (`worker` or `gyne`).
`Life stage`	Developmental stage of the individual (`pupa`, `adult`, etc.).
`Acon-1` to `Agp-1`	Allozyme genotype data at each locus, in Genepop 3-digit allele format.

Reprod mated P qn zyme genotypes Travis Co 1996

Allozyme genotypes for mated nestmate reproductive queens compiled from a study of a polygyne population in Texas (Travis Co.) conducted by Ross et al. (1996). These genotypes were used to estimate average pairwise nestmate queen relatedness (see Supporting Information for Hale Walker et al. [2025]). The genotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele.

Column Name	Description
`nest`	Identifier for the colony/nest.
`Aco-1` to `Pgm-1`	Allozyme genotype data at each locus in Genepop 3-digit allele format.
`mated?`	Indicates whether the queen was confirmed as mated (`yes`).

alate gyne zyme genotypes-GA M cols

Allozyme genotypes compiled from studies of monogyne colonies in northern Georgia (Morgan, Oconee, and Walton Counties) conducted by Ross et al. (1996) and Ross and Shoemaker (1997). The genotypes are from winged adult gynes and were used to estimate average pairwise nestmate relatedness (Hale Walker et al., 2025). The genotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele.

Column Name	Description
`Colony`	Identifier for the monogyne colony.
`Pgm-1` to `Aco-5`	Allozyme genotype data at each locus in Genepop 3-digit allele format.

Alate gyne msat genotypes-Ascunce GA M cols

Microsatellite genotypes for adult gynes (Ascunce series; single individual per nest) or mated reproductive queens (FU series) compiled from the study of Ascunce et al. (2011) and from unpublished data (K. G. Ross) for the monogyne social form in northern Georgia. These genotypes were used to estimate the inbreeding coefficient (FIS) (see Supporting Information Table S2 for Hale Walker et al. [2025]). The genotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele.

Column Name	Description
`Population`	Population source (e.g., northern Georgia monogyne).
`Individual`	Unique ID of the gyne (e.g., Ascunce-1, FU-2).
`red_ant` to `i_120`	Microsatellite genotype data at each locus in Genepop 3-digit format.

SE US monogyne nuc genotypes

Allozyme and microsatellite genotypes for adult gynes (single individual per nest; n=503 nests) compiled from a study of six monogyne populations in the southern U.S. conducted by Shoemaker et al. (1996). These genotypes were used to estimate the inbreeding coefficient (FIS) for the monogyne social form in this invasive range (see Supporting Information Table S2 for Hale Walker et al. [2025]). The genotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele.

Column Name	Description
`Population`	Geographic and social form origin of the sample (e.g., Georgia monogyne).
`Individual`	Unique ID of each sampled gyne.
`Pgm-1` to `Ddh-1`	Allozyme genotype data in Genepop 3-digit allele format.
`Sol-6` to `Sol-55`	Microsatellite genotype data in Genepop 3-digit allele format.

SE US polygyne nuc genotypes

Allozyme and microsatellite genotypes for adult gynes or mated reproductive queens (single individual per nest; n=557 nests) compiled from a study of six polygyne populations in the southern U.S. conducted by Shoemaker et al. (1996). These genotypes were used to estimate the inbreeding coefficient (FIS) for the polygyne social form in this invasive range (see Supporting Information Table S4 for Hale Walker et al. [2025]). The genotypes are in the Genepop 3-digit allele format (Raymond & Rousset, 1995; Rousset, 2008), with "000" indicating a missing allele.

Column Name	Description
`Population`	Geographic and social form origin of the sample (e.g., Georgia polygyne).
`Individual`	Unique ID of each sampled gyne or mated queen.
`Pgm-1` to `Ddh-1`	Allozyme genotype data in Genepop 3-digit allele format.
`Sol-6` to `Sol-55`	Microsatellite genotype data in Genepop 3-digit allele format.

newly mated qns Fis-Fisher dissert north GA

Estimate of the inbreeding coefficient (FIS) from newly mated monogyne queens captured at one site in northern Georgia (Oglethorpe Co.) on one day (see Supporting Information Table S2 for Hale Walker et al. [2025]). Hobs and Hexp were calculated for each locus using Genepop on the Web (v.4.0.10) (Raymond & Rousset, 1995; Rousset, 2008) and appear in the unpublished Ph.D. dissertation of M. A. Fisher (2013). Single-locus FIS values were estimated from them using Eq 7.7c in Hedrick (2005), with each single-locus estimate weighted by sample size. The overall estimate and its 95% confidence interval were obtained from 1000 bootstrap replicates over the 105 single-locus estimates.

Column Name	Description
`Marker type`	Type of genetic marker (e.g., microsatellite).
`Locus`	Name of the genetic locus analyzed.
`N (individuals)`	Number of individuals genotyped at the locus.
`Hobs`	Observed heterozygosity.
`Hexp`	Expected heterozygosity.
`Fis_single locus`	FIS (inbreeding coefficient) calculated for each locus.
`Fis_overall`	Overall FIS estimate and confidence interval (present only in select rows).

R script

`remating_analysis.R`

Provides the R (version 4.4.1) codes associated with Method S8: Analyses of remating probabilities. This R script estimates the probabilities of remating in polygyne fire ant queens with different male genotypes (SB vs. Sb). It compares observed mating-type frequencies to theoretical expectations under two models of remating probability. The script is self-contained and does not require any data file imports.

Purpose

To infer the likelihood that a queen remates with SB or Sb males, based on population-level observations from Florida and Georgia. It computes:

Best-fit remating probabilities
Estimated mating-type frequencies
Heatmaps of inferred error

Required R Packages

library(ggplot2)
library(scales)
library(magick)

How to Use

These data files are provided to support analyses of breeding systems and reproductive competition in polygyne fire ant colonies. Users may:

Reproduce key analyses and figures in the manuscript
Conduct additional analyses of queen longevity, reproductive skew, or genetic relatedness

Contact

For questions regarding the data or manuscript, please contact:

Haolin (Horace) Zeng – haolinzeng@uga.edu
Kenneth Ross – kenross@uga.edu

ORCID IDs

Kip Lacy: https://orcid.org/0000-0002-3149-8927
Haolin Zeng: https://orcid.org/0000-0002-9921-7335

References

Hale Walker S, Lacy KD, Ross KG, Zeng H. 2025. Reproductive competition in multiple-queen fire ant colonies: Insights from analyses of breeding systems. Molecular Ecology.
Raymond M, Rousset F. 1995. GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. Journal of Heredity 86:248-249.
Rousset F. 2008. Genepop'007: a complete reimplementation of the Genepop software for Windows and Linux. Molecular Ecology Resources 8:103-106.
Ross KG. 1993. The breeding system of the fire ant Solenopsis invicta: Effects on colony genetic structure. The American Naturalist 141:554–576.
Ross KG, Vargo EL, Keller L. 1996. Social evolution in a new environment: the case of introduced fire ants. Proceedings of the National Academy of Sciences, USA 93:3021-3025.
Ross KG, Shoemaker DD. 1997. Nuclear and mitochondrial genetic structure in two social forms of the fire ant Solenopsis invicta. Evolution 51:913–918.
Ross KG, Shoemaker DD. 2018. Unexpected patterns of segregation distortion at a selfish supergene in the fire ant Solenopsis invicta. BMC Genetics 19:101.