The dual strategy: Resisting and tolerating pathogens in ant colonies
Data files
Jun 09, 2026 version files 253.68 KB
-
Excel_Main_Table_Consumption_The_Dual_Strategy.csv
1.43 KB
-
Excel_Main_Table_Grooming_The_Dual_Strategy.csv
10.82 KB
-
Excel_Main_Table_Mortality_Production_The_Dual_Strategy.csv
1.92 KB
-
Excel_Main_Table_Necrophoresis_The_Dual_Strategy.csv
2.44 KB
-
Excel_Main_Table_Recruitment_The_Dual_Strategy.csv
69.91 KB
-
R_script_The_Dual_Strategy__Resisting_and_Tolerating_Pathogens_in_Ant_Colonies_Avanzi___Detrain_save_v2.R
79.84 KB
-
README.md
28.50 KB
-
Tables_for_R_The_Dual_Strategy_Resisting_and_Tolerating_Pathogens_in_Ant_Colonies.zip
58.83 KB
Jun 09, 2026 version files 253.68 KB
-
Excel_Main_Table_Consumption_The_Dual_Strategy.csv
1.43 KB
-
Excel_Main_Table_Grooming_The_Dual_Strategy.csv
10.82 KB
-
Excel_Main_Table_Mortality_Production_The_Dual_Strategy.csv
1.92 KB
-
Excel_Main_Table_Necrophoresis_The_Dual_Strategy.csv
2.44 KB
-
Excel_Main_Table_Recruitment_The_Dual_Strategy.csv
69.91 KB
-
R_script_The_Dual_Strategy__Resisting_and_Tolerating_Pathogens_in_Ant_Colonies_Avanzi___Detrain_save_v2.R
79.84 KB
-
README.md
28.50 KB
-
Tables_for_R_The_Dual_Strategy_Resisting_and_Tolerating_Pathogens_in_Ant_Colonies.zip
58.83 KB
Abstract
Ant colonies are vulnerable to disease outbreaks due to their dense, genetically related populations and confined living conditions. To mitigate the impact of entomopathogens such as Beauveria bassiana, ants employ resistance strategies to reduce pathogen load and tolerance strategies to endure infections without reducing pathogen presence. Investment in these strategies likely depends on the intensity and duration of pathogen exposure, and may be colony-specific and shaped by trade-offs. To investigate colony-specificity of sanitary responses, we divided Myrmica rubra ant nests into two sister colonies exposed to either a fixed or removable pathogen load, where workers were forced to respectively coexist with nestmate cadavers or could discard them. While sister colonies exhibited similar performance in tolerance-related behaviours, they differed in resistance responses. Under severe pathogenic risk (sporulating cadavers), workers engaged in more frequent social and extensive grooming, and colonies compensated for worker mortality by increasing brood production. A trade-off between tolerance and resistance strategies emerged during control but disappeared when colonies faced sporulating cadavers. These findings highlight the flexibility of ant colonies in adjusting their defence strategies and suggest that severe pathogenic threats may override constraints on resource allocation between tolerance and resistance strategies.
Dataset DOI: 10.5061/dryad.cjsxksnhs
Description of the data and file structure
The dataset contains seven files:
- The 'R script' file contains the R script used to perform the statistical analysis for the paper.
- The 'Tables for R' file contains the .txt tables necessary to run the R script.
The five 'Excel Tables' contain the raw data tables for various measurements:
- Excel_Main_Table_Grooming_The_Dual_Strategy.csv
- Excel_Main_Table_Mortality_Production_The_Dual_Strategy.csv
- Excel_Main_Table_Necrophoresis_The_Dual_Strategy.csv
- Excel_Main_Table_Recruitment_The_Dual_Strategy.csv
- Excel_Main_Table_Consumption_The_Dual_Strategy.csv
All CSV files are semicolon-delimited. Commas are used as decimal separators in all files except Excel_Main_Table_Recruitment_The_Dual_Strategy.csv, which uses periods/full stops.
Files and variables
File: R_script_The_Dual_Strategy__Resisting_and_Tolerating_Pathogens_in_Ant_Colonies_Avanzi___Detrain_save_v2.R
Description: The R script, combined with the R tables file, is ready to use and will guide you through the analysis of the data and the production of the figures presented in the paper and generated in R. The script begins with general information, followed by an index table to guide you through the statistical analysis.
File: Tables_for_R_The_Dual_Strategy_Resisting_and_Tolerating_Pathogens_in_Ant_Colonies.zip
Description: To run the script, all the .txt tables in the 'R tables' file must be in your R directory. The script was created using RStudio Desktop v1.3.1056 or more recent.
File: Excel_Main_Table_Grooming_The_Dual_Strategy.csv
Description: This data table summarises the behavioural observations of sanitary behaviours at various time points inside ant nests.
File: Excel_Main_Table_Mortality_Production_The_Dual_Strategy.csv
Description: This data table summarises the daily demographic assessment of the ant's nest conducted throughout the experiment.
File: Excel_Main_Table_Necrophoresis_The_Dual_Strategy.csv
Description: This data table summarises the necrophoric measurements taken after uninfected and sporulating cadavers were introduced into the ant nests on day 1 and 7.
File: Excel_Main_Table_Recruitment_The_Dual_Strategy.csv
Description: This data table summarises the data on colonial foraging efforts obtained during the recruitment experiments performed under control and infection conditions.
File: Excel_Main_Table_Consumption_The_Dual_Strategy.csv
Description: This data table summarises the food consumption data obtained during the recruitment experiments performed under control and infection conditions.
Glossary of variables used in each of the five .csv tables
Note: Please refer to the glossary for each table separately, as the same variable name can have different meanings depending on the data table. For example, 'ID' in the necrophoresis table refers to cadaver identification number, while in the mortality table it refers to a worker identification number.
1) Glossary for the 'Excel_Main_Table_Grooming_The_Dual_Strategy.csv' table:
- Colony: Colony number.
- Condition: Condition of exposure to cadaver.
- Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Removable: Colony in the removable condition where cadavers were free to be discarded
- Time: A categorical list of the time points at which the 3-minute video recordings started.
- J1_9h30: Video recording started on day 1 at 9h30
- J1_11h30: Video recording started on day 1 at 11h30
- J1_12h30: Video recording started on day 1 at 12h30
- J2_12h: Video recording started on day 2 at 12h
- J7_9h30: Video recording started on day 7 at 9h30
- J7_11h30: Video recording started on day 7 at 11h30
- J15_12h30: Video recording started on day 7 at 12h30 (typing error in the table, should have been J7)
- J8_12h: Video recording started on day 8 at 12h
- J15_12h: Video recording started on day 15 at 12h
- AlloGrooming: The number of events in the nest where ant workers perform 'allo-grooming' behaviour.
- SelfGrooming: The number of events in the nest where ant workers perform 'self-grooming' behaviour.
- ExtensiveGrooming: The number of events in the nest where ant workers perform 'extensive-grooming' behaviour.
- Brood: The number of events in the nest where ant workers perform 'brood care' behaviour.
- Density: Average number of ant workers in the nest.
- AlloPerDensity = AlloGrooming / Density.
- SelfGPerDensity = SelfGrooming / Density.
- ExtensPerDensity = ExtensiveGrooming / Density.
- BroodPerDensity = Brood / Density.
2) Glossary for the 'Excel_Main_Table_Mortality_Production_The_Dual_Strategy.csv' table:
- Colony: Colony number.
- Condition: Condition of exposure to cadaver.
- Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Removable: Colony in the removable condition where cadavers were free to be discarded
- Period: Categorical Period of exposure to cadaver, each lasting 1 week.
- Low_risk: Control week where ant nest was exposed to uninfected cadavers
- High_risk1st: First week of infection where ant nest was exposed to sporulating cadavers
- High_risk2nd: Second week of infection where ant nest was exposed to sporulating cadavers
- Morality_rate: The number of workers dying per day that later sporulated.
- Brood_Mortality_rate: The number of larvae/nymphae dying per day that later sporulated.
- Brood_Production_rate: The number of brood packs laid by the queens per day, with a brood pack being defined as an area of 1 mm² of the nest covered by eggs.
- Brut_Morality_rate: The number of workers dying per day
- Brut_Brood_Mortality_rate: The number of larvae/nymphae dying per day
3) Glossary for the 'Excel_Main_Table_Necrophoresis_The_Dual_Strategy.csv' table:
- ID: Identification of each cadaver analyzed during the experiment
- Colony: Colony number.
- Risk: Types of cadavers dropped inside the ant nest.
- Low = Uninfected cadavers
- High = Sporulating cadavers
- Status: The status of the cadaver at the end of the experiment.
- 0 = not discarded
- 1 = discarded
- Time_of_Removal: The time at which the cadaver was discarded outside the nest.
4) Glossary for the 'Excel_Main_Table_Recruitment_The_Dual_Strategy.csv' table:
- Time: The minute of the recruitment experiment at which the measurement was taken.
- Group: Categorical list of condition and exposure to cadavers.
- Fixed_Uninfected: Colonies in the fixed cadaver condition exposed to uninfected corpses
- Fixed_Sporulating: Colonies in the fixed cadaver condition exposed to Sporulating corpses
- Removable_Uninfected: Colonies in the Removable cadaver condition exposed to uninfected corpses
- Removable_Sporulating: Colonies in the Removable cadaver condition exposed to Sporulating corpses
- SubGroup: Categorical list of cadaver exposure.
- Uninfected: Colonies exposed to Uninfected corpses
- Sporulating: Colonies exposed to Sporulating corpses
- Colony: Colony number.
- Sucre: Number of workers at the sugar food source.
- Protein: Number of workers at the protein food source.
- Sum: Total number of workers at the protein food source (Sucre+Protein).
- N_WorkerInColony: Number of living workers in the colony at that time.
- Percentage_of_Colony: Percentage of the workforce at the food sources.
- Mean_Smoothed: The average percentage of the workforce at food sources, smoothed on three values.
5) Glossary for the 'Excel_Main_Table_Consumption_The_Dual_Strategy.csv' table:
- Colony: Colony number.
- Condition: Condition of exposure to cadaver.
- Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Removable: Colony in the removable condition where cadavers were free to be discarded
- Exposure: Types of cadavers dropped inside the ant nest.
- Low = Uninfected cadavers
- High = Sporulating cadavers
- Group: Categorical list of condition and exposure to cadavers.
- Fixed_Low: Colonies in the fixed cadaver condition exposed to uninfected corpses
- Fixed_High: Colonies in the fixed cadaver condition exposed to Sporulating corpses
- Removable_Low: Colonies in the Removable cadaver condition exposed to uninfected corpses
- Removable_High: Colonies in the Removable cadaver condition exposed to Sporulating corpses
- Total: Total consumption by the colony during the recruitment experiment in grams.
- Mealworms: Mealworms (protein) consumption by the colony during the recruitment experiment in grams.
- Sugar: Sugar consumption by the colony during the recruitment experiment in grams.
Glossary of variables used in each of the 16 .txt tables used to run the R_script on R:
Note: Please refer to the glossary for each table separately, as the same variable name can have different meanings depending on the data table. For example, 'ID' in the necrophoresis table refers to cadaver identification number, while in the mortality table it refers to a worker identification number.
1) Glossary for 'Consomation.txt':
-
Mother_colony: Colony number.
-
Condition: Condition of exposure to cadaver.
- Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Removable: Colony in the removable condition where cadavers were free to be discarded
-
Exposure: Types of cadavers dropped inside the ant nest.
- Low = Uninfected cadavers
- High = Sporulating cadavers
-
Group: Categorical list of condition and exposure to cadavers.
- Fixed_Low: Colonies in the fixed cadaver condition exposed to uninfected corpses
- Fixed_High: Colonies in the fixed cadaver condition exposed to Sporulating corpses
- Removable_Low: Colonies in the Removable cadaver condition exposed to uninfected corpses
- Removable_High: Colonies in the Removable cadaver condition exposed to Sporulating corpses
-
Total: Total consumption by the colony during the recruitment experiment in grams.
-
Mealworms: Mealworms (protein) consumption by the colony during the recruitment experiment in grams.
-
Sugar: Sugar consumption by the colony during the recruitment experiment in grams.
2) Glossary for 'data_mortality.txt':
-
ID: Each ant worker involved in the experiment is identified by a ID number. The first number is the colony number, followed by either R or T for the condition, and then the number of the ant within its colony, from 1 to 400.
-
Colony: Colony number.
-
Condition: Condition of exposure to cadaver.
- Tolerance = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Resistance = Removable: Colony in the removable condition where cadavers were free to be discarded
-
Status: Survival status.
- Dead = 1: The individual died during the experiment
- Alive = 0: The individual survived until the end of the experiment
-
DayOfDeath: The day of the experiment when the 'Status' occurred. If alive, this is simply the last day of the experiment ('26').
-
Activity_level: relic from behavioural analysis where colonies were categorized by activity levels.
3) Glossary for 'dataCor3.txt':
- Colony: Colony number.
- Condition: Condition of exposure to cadaver.
- Tolerance = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Resistance = Removable: Colony in the removable condition where cadavers were free to be discarded
- Time = Exposure period: Types of cadavers dropped inside the ant nest.
- Ctrl = Control = period of experiment with Uninfected cadavers
- Inf = Infection = period of experiment with Sporulating cadavers
- Activity: This is a relic from behavioural analysis, where colonies were categorised by activity levels. Refer to the raw activity levels.
- Activity_Percapita: This is a relic from behavioural analysis, where colonies were categorised by activity levels. Refer to the raw activity levels divided per 'Density'.
- Morality: The number of workers dying per day that later sporulated.
- Brood_Morta: The number of larvae/nymphae dying per day that later sporulated.
- Morta_brut: The number of workers dying per day.
- Morta_brood_Brut: The number of larvae/nymphae dying per day.
- Brood_prod: The number of brood packs laid by the queens per day, with a brood pack being defined as an area of 1 mm² of the nest covered by eggs.
- Surface_oeuf: Area covered by egg pack in mm².
- Ind_to_food_source: Number of workers present at food sources.
- Recruitment: Percentage of the workforce present at the food sources.
- Total_conso: Total consumption by the colony during the recruitment experiment in grams.
- Mealworms: Mealworms (protein) consumption by the colony during the recruitment experiment in grams.
- Sugar: Sugar consumption by the colony during the recruitment experiment in grams.
- Brood_Care: The average number of events in the nest where ant workers perform 'brood care' behaviour.
- Brood_Care_Percapita: Brood_Care / Number of workers in the nest.
- AlloG: The average number of events in the nest where ant workers perform 'Allo-grooming' behaviour.
- AlloG_Percapita: AlloG / Number of workers in the nest.
- SelfG: The average number of events in the nest where ant workers perform 'Self-grooming' behaviour.
- SelfG_Percapita: SelfG / Number of workers in the nest.
- ExtenG: The average number of events in the nest where ant workers perform 'Extensif-grooming' behaviour.
- ExtenG_Percapita: ExtenG / Number of workers in the nest.
- Necrophoresis: Average time the colony took to discard a cadaver from the nest (in hours).
- 1onNecrophoresis: 1 / Average time the colony took to discard a cadaver from the nest.
- Sanitarization_time: Time the colony took to discard every cadaver from the nest (in hours).
4) Glossary for 'DataPerPeriod.txt':
- Colony: Colony number.
- Treatment: Condition of exposure to cadaver.
- Tolerance = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid.
- Resistance = Removable: Colony in the removable condition where cadavers were free to be discarded.
- Period: Categorical Period of exposure to cadaver, each lasting 1 week.
- Control: Control week where ant nest was exposed to uninfected cadavers
- Treatment: First week of infection where ant nest was exposed to sporulating cadavers
- End: Second week of infection where ant nest was exposed to sporulating cadavers
- Morality_rate: The number of workers dying per day that later sporulated.
- Brood_Mortality_rate: The number of larvae/nymphae dying per day that later sporulated.
- Brood_Production_rate: The number of brood packs laid by the queens per day, with a brood pack being defined as an area of 1 mm² of the nest covered by eggs.
- Brut_Morality_rate: The number of workers dying per day.
- Brut_Brood_Mortality_rate: The number of larvae/nymphae dying per day.
5) Glossary for 'DataRecru2.txt':
- Time: The minute of the recruitment experiment at which the measurement was taken.
- Group: Categorical list of condition and exposure to cadavers.
- Free_Uninfected: Colonies in the fixed cadaver condition exposed to uninfected corpses
- Free_Sporulating: Colonies in the fixed cadaver condition exposed to Sporulating corpses
- Locked_Uninfected: Colonies in the Removable cadaver condition exposed to uninfected corpses
- Locked_Sporulating: Colonies in the Removable cadaver condition exposed to Sporulating corpses
- SubGroup: Categorical list of cadaver exposure.
- Uninfected: Colonies exposed to Uninfected corpses
- Sporulating: Colonies exposed to Sporulating corpses
- Colony: Colony number.
- Sucre: Number of workers at the sugar food source.
- Protein: Number of workers at the protein food source.
- Sum: Total number of workers at the protein food source (Sucre+Protein).
- N_WorkerInColony: Number of living workers in the colony at that time.
- Percentage_of_Colony: Percentage of the workforce at the food sources.
- Mean_Smoothed: The average percentage of the workforce at food sources, smoothed on three values.
6) Glossary for the 3 tables 'dataScoreFree2.txt' & 'dataScoreLock2.txt' & 'dataScoreTot2.txt':
- ID: Colony number.
- Condition: Condition of exposure to cadaver.
- Locked = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Free = Removable: Colony in the removable condition where cadavers were free to be discarded
- Score_Tol_Ctrl: The average tolerance score obtained by the colonies in the fixed condition during the control.
- Score_Tol_Inf: The average tolerance score obtained by the colonies in the fixed condition during the Infection.
- Score_Res_Ctrl: The average resistance score obtained by the colonies in the fixed condition during the control.
- Score_Res_Inf: The average resistance score obtained by the colonies in the fixed condition during the infection.
Note on scores:
For each experimental colony, we computed overall scores reflecting its investment in resistance- or tolerance-related strategies, calculated separately for control and infection periods. The resistance score was derived from four proxies: Allo-grooming events per capita, Extensive-grooming events per capita, Self-grooming events per capita, and Sanitarization_time (the latter applicable only for colonies under removable cadaver condition). For each proxy, the 14 sister colonies were ranked and assigned points from 1 (lowest value) to 14 (highest value). Resistance scores were obtained by summing points across all four resistance-related proxies and then normalizing them between 0 and 1, where 1 represented the maximum possible score (14 points × 4 proxies = 56 points) and 0 the minimum (1 point × 4 proxies = 4 points). In the fixed cadaver condition, where necrophoresis could not be measured, the maximum possible score was adjusted to 42 points. The same procedure was applied to tolerance-related strategies, using two recruitment proxies (worker mobilization to food sources and food consumption) and two brood demographic proxies (egg-laying rate and the inverse of brood mortality rate). Tolerance scores were calculated by summing the ranked points across these proxies.
7) Glossary for the 3 tables 'dataTradeoffFree.txt' & 'dataTradeoffLocked.txt' & 'dataTradeoffTotal.txt':
- ID: Colony number.
- Condition: Condition of exposure to cadaver.
- Locked = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Free = Removable: Colony in the removable condition where cadavers were free to be discarded
- Consumption_Ctrl: The total amount of food consumed (in grams) during the recruitment experiment under control conditions (uninfected cadavers).
- Consumption_Inf: The total amount of food consumed (in grams) during the recruitment experiment in infection conditions (sporulating cadavers).
- Recruitment_Ctrl: The average percentage of the workforce at food sources during the recruitment experiment under control conditions (uninfected cadavers).
- Recruitment_Inf: The average percentage of the workforce at food sources during the recruitment experiment in infection conditions (sporulating cadavers).
- Egg_surface_Ctrl: Surface cover by egg pack (in mm²) at the end of control conditions (uninfected cadavers).
- Egg_surface_Inf: Surface cover by egg pack (in mm²) at the end of infection conditions (sporulating cadavers).
- Brood_morta_brut_Ctrl: The average number of nymphae/larvae dying per day during the control period (uninfected cadavers).
- Brood_morta_brut_Inf: The average number of nymphae/larvae dying per day in infection conditions (sporulating cadavers).
- ExtG_perCapita_Ctrl: The average number of events in the nest where ant workers perform 'extensive-grooming' behaviour divided by the density in the nest at that time, under control conditions (uninfected cadavers).
- ExtG_perCapita_Inf: The average number of events in the nest where ant workers perform 'extensive-grooming' behaviour divided per the density in the nest at that time, in infection conditions (sporulating cadavers).
- AlloG_perCapita_Ctrl: The average number of events in the nest where ant workers perform 'allo-grooming' behaviour divided by the density in the nest at that time, under control conditions (uninfected cadavers).
- AlloG_perCapita_Inf: The average number of events in the nest where ant workers perform 'allo-grooming' behaviour divided by the density in the nest at that time, in infection conditions (sporulating cadavers).
- SelfG_perCapita_Ctrl: The average number of events in the nest where ant workers perform 'self-grooming' behaviour divided by the density in the nest at that time, under control conditions (uninfected cadavers).
- SelG_perCapita_Inf: The average number of events in the nest where ant workers perform 'self-grooming' behaviour divided by the density in the nest at that time, in infection conditions (sporulating cadavers).
- Inv_Necrophoresis_Ctrl: 1 / Average time the colony took to discard an uninfected cadaver from the nest
- Inv_Necrophoresis_Inf: 1 / Average time the colony took to discard a sporulating cadaver from the nest
- Egg_laying_Ctrl: The average number of brood packs laid by the queens per day during the control period, with a brood pack being defined as an area of 1 mm² of the nest covered by eggs.
- Egg_laying_Inf: The average number of brood packs laid by the queens per day during the infection period, with a brood pack being defined as an area of 1 mm² of the nest covered by eggs.
- Broodcare_perCapita_Ctrl: The average number of events in the nest where ant workers perform 'brood care' behaviour divided by the density in the nest at that time, under control conditions (uninfected cadavers).
- Broodcare_perCapita_Inf: The average number of events in the nest where ant workers perform 'brood care' behaviour divided by the density in the nest at that time, in infection conditions (sporulating cadavers).
- Morta_Worker_Ctrl: The average number of workers dying per day during the control period (uninfected cadavers).
- Morta_Worker_Inf: The average number of workers dying per day during the infection period (sporulating cadavers).
8) Glossary for 'FitnessNorm.txt':
- Colony: Colony number.
- ConditionOld: Condition of exposure to cadaver.
- Locked = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Free = Removable: Colony in the removable condition where cadavers were free to be discarded
- Load: Categorical list of cadaver exposure.
- 1Low = Uninfected: Colonies exposed to Uninfected corpses
- 2High = Sporulating: Colonies exposed to Sporulating corpses
- Condition: Categorical list of 'ConditionOld' + 'Load'.
- Removable_1Low: Colonies in the Removable cadaver condition exposed to uninfected corpses
- Removable_2High: Colonies in the Removable cadaver condition exposed to Sporulating corpses
- Fixed_1Low: Colonies in the fixed cadaver condition exposed to uninfected corpses
- Fixed_2High: Colonies in the fixed cadaver condition exposed to Sporulating corpses
- Morta_brood_Brut: The number of larvae/nymphae dying per day.
- Morta_brut: The number of worker dying per day.
- Brood_Mortav1: It is a relic of the number of larvae/nymphs dying per day that later sporulated, with an error in the calculation.
- Worker_Mortav1: It is a relic of the number of worker dying per day that later sporulated, with an error in the calculation.
- Brood_Mortav2: The number of larvae/nymphae dying per day that later sporulated.
- Worker_Mortav2: The number of workere dying per day that later sporulated.
9) Glossary for 'LogRankActivityBrood.txt':
- ID: Each brood used in the experiment is identified by an ID number. The first number is the colony number, followed by either R (Resistance) or T (Tolerance) for the condition and then the number of the ant within its colony, from 1 to 100.
- Colony: Colony number.
- Condition: Condition of exposure to cadaver.
- Tolerance = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid.
- Resistance = Removable: Colony in the removable condition where cadavers were free to be discarded.
- Status: Survival status.
- Dead = 1: The individual died during the experiment
- Alive = 0: The individual survived until the end of the experiment
- DayOfDeath: The day of the experiment when the 'Status' occurred. If alive, this is simply the last day of the experiment ('26').
- Activity_level: relic from behavioural analysis where colonies were categorized by activity levels.
10) Glossary for 'LogRankNecro.txt':
- ID: Each cadaver used in the experiment is identified by an ID number. The first number is the colony number, followed by either R (Resistance) or T (Tolerance) for the condition and then the number of the ant within its colony, from 1 to 20. 1 to 10 being uninfected cadavers and 11 to 20 being sporulating cadavers.
- Colony: Colony number.
- Risk: Categorical list of cadaver exposure.
- Low = Uninfected: Colonies exposed to Uninfected corpses
- High = Sporulating: Colonies exposed to Sporulating corpses
- Status: Survival status.
- Dead = 1: The individual died during the experiment
- Alive = 0: The individual survived until the end of the experiment
- DayOfDeath: The time at which the cadaver was discarded from the nest (in min).
- Activity_level: relic from behavioural analysis where colonies were categorized by activity levels
11) Glossary for 'MortavsEggLv2.txt':
- ID: Colony number.
- Egg_laying_Low: The average number of brood packs laid by the queens per day during the control period, with a brood pack being defined as an area of 1 mm² of the nest covered by eggs.
- Egg_laying_High: The average number of brood packs laid by the queens per day during the infection period, with a brood pack being defined as an area of 1 mm² of the nest covered by eggs.
- Brood_morta_Low: The average number of nymphae/larvae dying per day during the control period.
- Brood_morta_High: The average number of nymphae/larvae dying per day during the infection period.
- Morta_Worker_Low: The average number of workers dying per day during the control period.
- Morta_Worker_High: The average number of workers dying per day during the infection period.
12) Glossary for 'simplified_data_summarized.txt':
- Colony: Colony number.
- Condition: Condition of exposure to cadaver.
- Tolerance = Fixed: Colony in the fixed condition where cadavers were locked inside a metallic mesh grid
- Resistance = Removable: Colony in the removable condition where cadavers were free to be discarded
- Time: A categorical list of the time points at which the 3-minute video recordings started.
- J2_9h30: Video recording started on day 1 at 9h30
- J2_11h30: Video recording started on day 1 at 11h30
- J2_12h30: Video recording started on day 1 at 12h30
- J3_12h: Video recording started on day 2 at 12h
- J5_9h30: Video recording started on day 7 at 9h30
- J5_11h30: Video recording started on day 7 at 11h30
- J5_12h30: Video recording started on day 15 at 12h30
- J6_12h: Video recording started on day 8 at 12h
- AlloGrooming: The number of events in the nest where ant workers perform 'allo-grooming' behaviour.
- SelfG: The number of events in the nest where ant workers perform 'self-grooming' behaviour.
- Extensive: The number of events in the nest where ant workers perform 'extensive-grooming' behaviour.
- Brood: The number of events in the nest where ant workers perform 'brood care' behaviour.
- Density: Average number of ant workers in the nest
- Activity: relic activity number, was not used in the analysis
- AlloPerDensity = AlloGrooming / Density
- SelfGPerDensity = SelfGrooming / Density
- ExtensPerDensity = ExtensiveGrooming / Density
- BroodPerDensity = Brood / Density
- ActPerDensity: relic activity number / density, was not used in the analysis
Access information
Other publicly accessible locations of the data:
- N/A
Data was derived from the following sources:
- N/A
Biological models
The ant species Myrmica rubra is a common inhabitant of temperate regions across Europe, Russia, and central to northern Asia. It has also become an invasive species in North America, where it can form supercolonies. This species is both polygynous and polyandrous; colonies typically house between 1 and 20 queens, and each queen can mate with several unrelated males. Queens from the same nest may be related or unrelated, and workers within a colony often originate from mixed parentage. As a result of this polygyny and polyandry, the genetic structure of M. rubra colonies is highly variable, with an average relatedness between nestmates of around 0.26 ± 0.19 and between nests within a population approaching zero. M. rubra typically inhabits semi-open forests and grasslands, building nests in a variety of substrates such as soil beneath stones, rotting wood, or roots of nettles and bramble bushes. As an omnivorous species that frequently preys on arthropods, M. rubra workers are regularly exposed to insect remains, which can serve as sources of entomopathogens. The genus Myrmica encounters a broad range of parasites and is commonly exposed in nature to the generalist entomopathogenic fungus Beauveria bassiana. Although genetically related workers live together at high densities within confined nest sites, pathogen transmission rarely leads to disease outbreaks that threaten colony survival. As a result, M. rubra has become a well-established biological model for the study of social immunity. Infection by B. bassiana begins with the adhesion of conidia to the insect cuticle, followed by the release of an enzymatic cocktail that enables penetration of the host body. Once inside the host, the fungal mycelium grows and invades the hemocoel, ultimately causing host death. At the final stage of infection, the fungus produces conidial propagules that enable further transmission of B. bassiana pathogen and give the cadaver its characteristic fluffy appearance, commonly referred to as “white muscardine”.
Nest design and experimental conditions
Seven Myrmica rubra colonies were collected during the summers of 2020 and 2021 in woodland areas at Aiseau-Presles (Belgium, Province of Hainaut; 50°25.657′ N, 4°35.674′ E,) and identified using the identification key of Blatrix. Nests were excavated from the litter and decaying wood logs. Ant colonies were maintained in a climate-controlled room under constant temperature and humidity (21 ± 1°C and 50 ± 5% RH) and a 12 h:12 h dark:light photoperiod. Darkened, humidified test tubes covered with red filter paper were provided as temporary nesting sites.
From each of these seven mother colonies, two experimental sister colonies were established. Each sister colony consisted of two queens, 400 workers (approximately one-third external workers and two-thirds internal workers), and 100 larvae at the second and third instars. External workers were sampled from ants walking outside the nests in the mother colonies, whereas internal workers were sampled from nesting test tubes. Queens were selected based on the following criteria: (i) only macrogynes were used (microgynes were excluded); (ii) winged gynes were excluded; and (iii) queens were only collected from nest tubes containing brood.
Before the start of the experiments, ants were transferred to an experimental tray (figure 1a; 30 × 22 × 7.5 cm), the bottom of which was covered with a 1-cm-thick plaster layer, and the edges were coated with Fluon™ to prevent ants from escaping. A hole was drilled in the tray directly beneath the experimental nest. A cotton wick was inserted into the hole and placed over a sponge soaked daily with water in order to keep the plaster beneath the nest consistently moist. Ants were housed in a hexagon-shaped nest with a single chamber and a single entrance (figure 1bde). The nest was made of two square translucent red Plexiglas plates (3 mm thick). The 3-mm gap between these plates created a space in which ants could settle in a single layer, facilitating in-nest observations. The roof plate featured a central circular hole (10.7 mm in diameter) sealed with a circular plug. This hole was used to introduce ant cadavers into the nest. Below the roof opening and circular plug, each nest had a metallic mesh cage (figure 1c). The mesh size allowed workers to contact cadavers with their legs or antennae, while preventing ants from passing through or removing the cadavers.
Sister colonies were maintained under identical nesting conditions, except for cadaver removability. In one condition, ten Myrmica rubra nestmate cadavers were placed inside a metallic mesh cage and could not be removed by workers (fixed cadaver condition; figure 1e), whereas in the other condition, the absence of the mesh cage allowed workers to discard the ten cadavers from the nest (removable cadaver condition; figure 1d). This difference was implemented during cadaver introduction. In the removable cadaver condition, the roof plug and metallic mesh cage were gently removed, and ten cadavers were quickly introduced through the roof opening before the plug was replaced (figure 1d). The following morning at 09:00, once all cadavers had been discarded by workers, the empty metallic mesh cage was returned to its original position. In the fixed cadaver condition, the same procedure was followed, except that the metallic cage remained in place throughout the experiment, preventing any corpse removal (figure 1e).
Uninfected cadavers were dead nestmates originating from the same mother colony as the tested sister colonies. These ants were killed by freezing one day prior to the experiment. This delay ensured that 24 h had elapsed, allowing the cuticular profiles of dead workers to become enriched in oleic and linoleic acids. These chemical changes enable recognition of cadavers by workers and trigger necrophoresis. Sporulating cadavers were also nestmates from the mother colony but had died following infection by the entomopathogenic fungus Beauveria bassiana (strain Naturalis-LTM). The production of sporulating cadavers was completed prior to the experiment. A droplet of Naturalis-LTM insecticide (approximately 4 µL) was topically applied to the abdomen of nestmates from the mother colony, which were then isolated in Petri dishes containing moist cotton until death. On the day of death, cadavers were washed with ethanol and distilled water following Lacey’s method to prevent contamination by microorganisms other than B. bassiana. Cadavers were subsequently placed on moist filter paper in parafilm-sealed Petri dishes and kept in a temperature-controlled cabinet at 25°C until fungal sporulation covered the ant body.
Experimental Timeline, Measurements, and Statistical Analyses
To account for the high intercolonial variability of Myrmica rubra, we used a within-colony repeated-measures design in which each colony served as its own control. Colonies were first exposed to uninfected cadavers (control period) and subsequently to sporulating cadavers (infection period), thus maintaining constant colony composition and organisation across treatments. This approach minimized confounding colony-level differences and contributed to isolate the effects of infection on sanitary behaviours.
To assess social immunity, we characterised colony-specific traits at different time points, including fitness proxies, resistance-related behaviours and tolerance-related behaviours. Fitness proxies were quantified using brood and worker mortality, with brood mortality defined as the number of dead larvae and pupae. Resistance-related behaviours included necrophoresis and three types of grooming (self-grooming, allo-grooming and extensive-grooming). Tolerance-related behaviour included brood production, recruitment to food sources and food consumption. Statistical analyses corresponding to each measurement are described below and were conducted using RStudio (RStudio Desktop v1.3.1056). Graphs and tables were generated using Excel and R (ggplot2 and flexplot packages). The R code used to perform these analyses is provided in the electronic supplementary materials for the reader's reference.
Fitness proxies
Worker and brood survival were used as proxies for colony fitness. The experiment was divided into two periods (figure 2): a control period (Day 1 to 7), initiated by the introduction of ten uninfected cadavers into the nest, and an infection period (Day 7 to 21), initiated by the introduction of ten sporulating cadavers. These periods allowed us to assess the immediate and short-term effects of a routine sanitary condition and a severe fungal infection, respectively. Mortality was assessed daily at 10:00 by counting and removing dead larvae, pupae, and workers. Larvae and pupae were considered dead when they were rejected outside the nest. Dead individuals were removed from the foraging area, counted, cleaned following Lacey’s method, placed in moistened Petri dishes (2 cm in diameter) sealed with parafilm, and stored in a temperature-controlled cabinet at 25°C. Sporulation of those dead individuals was assessed one week post-mortem to determine whether death could be attributed to infection by B. bassiana. During the daily assessment, all living individuals were also counted to track population size. Counts of live individuals and recovered dead ones were cross-checked to minimize errors.
Mortality rates were compared between control and infection periods using paired Wilcoxon tests (wilcox.test function, stats package). Mortality rates between fixed and removable conditions were similarly compared using paired Wilcoxon tests. Additionally, because mortality served as an inverse proxy for fitness, a transformed fitness index is provided in the Supplementary Information (electronic supplementary material figure S3).
Resistance-related behaviours
The resistance level of each colony was assessed by measuring grooming events per capita and the speed of necrophoresis, defined as the average time required to discard cadavers from the nest. Grooming behaviour was recorded on Days 1, 2, 7, 8, and 15, and necrophoretic behaviour on Days 1 and 7 (figure 2). The nest interior was monitored using iSpy® software (v7.1.2.0) and a Logitech C920® webcam (1920 × 1080 pixels, 15 fps).
Three types of grooming behaviours were quantified: extensive grooming, self-grooming, and allo-grooming. Similarly to acidopore grooming used for sanitation by formicine ants, M. rubra performs a functionally similar but anatomically distinct behaviour, termed extensive grooming. As described by Pereira et al., extensive grooming is a behaviour characterised by bending of the gaster followed by abdominal tip-to-mouth contact to ingest poison gland secretion using the sting apparatus. In the present study, self-grooming was defined as any grooming behaviour other than extensive grooming and performed by a single ant. Allo-grooming referred to any grooming interaction involving at least two nestmates.
Three-minute video recordings of the nest interior were taken at multiple time points (in blue, figure 2), during which all grooming events were counted. Baseline recordings were collected on Days 1 and 7 at 09:30, prior to the introduction of uninfected and sporulating cadavers, respectively. Baseline values for the three grooming behaviours were similar between day 1 and day 7. Additional recordings were obtained 30 and 90 min after cadaver introduction, to assess early behavioural responses while minimising disturbance effects associated with the introduction procedure. A further recording was shot the following days at 12:00 captured medium-term responses, when cadavers had been removed in the removable treatment and behavioural adjustments to cadavers confined within the mesh cage were established in the fixed treatment. Finally, a last recording was shot on Day 15, one week after sporulating cadaver introduction, to evaluate the persistence of long-term changes in grooming behaviour. For each video, the number of workers inside the nest was counted at both the start and end of the 3-min recording and the mean was used to calculate grooming events per capita.
To investigate cadaver-induced changes, the number of grooming events per capita measured at baseline was subtracted from the ones measured at each time point after cadaver introduction. For each tested condition (removable or fixed cadavers), these Δ grooming levels were compared across time points by using Friedman tests (function friedman.test, package stats) followed, when significant, by a Nemenyi post hoc test (function frdAllPairsNemenyiTest, package PMCMRplus). Additionally, we tested whether grooming levels triggered by the insertion of cadavers deviated from the baseline values by using Wilcoxon test with µ = 0 (function wilcox.test, package stats). Effect sizes for Wilcoxon and Friedman tests were calculated using wilcox_effsize and friedman_effsize functions respectively (rstatix package).
To quantify necrophoresis in colonies under the removable cadaver condition, two additional uninterrupted five-hour video recordings were conducted immediately after the introduction of uninfected and sporulating cadavers (in purple, figure 2). Each of the ten introduced cadavers was tracked, and the time at which each of them was discarded from the nest was determined. After the last cadaver had been removed, all ten cadavers were removed from the foraging area to avoid confusion with naturally dead workers during the control period and to limit exposure to infectious material during the infection period. Discarding times were compared between control (uninfected cadavers) and infection (sporulating cadavers) periods within each colony using log-rank tests (survdiff function, survival package).
Tolerance-related behaviours
Tolerance was estimated using a recruitment experiment and a daily quantification of brood production by queens. Brood production was measured daily by estimating the number of newly laid egg packs, defined as a cluster covering approximately 1 mm². Brood production was calculated as the daily increase in the egg-covered area (mm²). On the final day of the experiment, nests were emptied of workers and the total number of egg packs was counted under a stereomicroscope to validate daily estimates. Average daily egg laying rates were compared between control and infection periods, and between fixed and removable conditions, using paired Wilcoxon tests (wilcox.test function, stats package). Effect sizes for Wilcoxon tests were calculated using wilcox_effsize function from the rstatix package. In order to assess whether ant colonies implemented compensatory strategies to mitigate pathogen-induced mortality, we correlated mortality rates and egg laying rates using Spearman’s rank test (function cor.test, package stats), calculated separately for control and infection periods.
Recruitment experiments were conducted to evaluate how exposure to sanitary risks influences foraging activity and food consumption in ant colonies, under the assumption that colonies would increase food intake and worker investment following infection, to offset the energetic costs associated with immune defences. Trials were conducted from 11:00 to 12:00 on Days 4 and 10, i.e. three days after the introduction of uninfected or sporulating cadavers, respectively (in orange, figure 2). Two days prior to each trial, sugar and protein sources were removed from the foraging area to induce mild starvation, which was sufficient to trigger recruitment in preliminary tests. At the start of each trial, two food sources were simultaneously placed in the foraging area, opposite the nest and 4 cm from the tray edge. These food sources consisted respectively of (i) a 0.3 M sugar solution prepared from white and brown commercial sugar in a 2:1 ratio, and (ii) internal tissues of Tenebrio molitor mealworms extracted from the cuticle and placed in an empty food source. Those two food sources were 3cm diameter white cross shaped feeders, similar to the one in which protein was provided ad libitum (fig1a.). The number of workers present at both food sources was visually counted every two minutes for one hour. Food sources were weighed immediately before and after each trial using a precision balance (Kern ABT 120-5DNM™, 0.1 mg precision). To correct for evaporation, two identical food sources were placed outside the experimental trays and weighed following the same procedure.
Recruitment was quantified as the average percentage of the worker population present at the two food sources, recorded every two minutes over the course of the one-hour experiment. To assess the effects of experimental conditions and time on recruitment, a Generalized Linear Model (GLM) was used. The model selection process, including comparisons of AIC and BIC values, is detailed in the electronic supplementary material (supplementary material table S3). The additive model (Experimental condition + time) was selected with ‘condition’ predictor referring to the four treatment groups: control removable cadavers, control fixed cadavers, infection removable cadavers, and infection fixed cadavers. Differences in total food consumption were tested using Kruskal-Wallis tests (kruskal.test function, stats package). Two separate tests were conducted: one comparing exposure levels (control vs. sporulating), and another comparing cadaver removability conditions (fixed vs. removable cadavers).
Colony-specificity of sanitary responses
To test for a colonial specificity of sanitary responses, we analysed whether sister colonies originating from the same mother colony exhibited correlated responses for each sanitary proxy. Correlations were assessed by comparing each proxy measured in a colony exposed to the fixed cadaver condition with that of its paired sister colony exposed to the removable condition (cor.test function, stats package). Because brood mortality did not meet the assumptions of parametric tests, Spearman rank correlations were used instead (function cor.test, package stats). Analyses were conducted separately for the control and infection periods, during which colonies were exposed to uninfected or sporulating cadavers, respectively. For each colony and period, we correlated the measurements used to quantify colonial fitness, tolerance, or resistance, retaining more direct metrics when necessary to ensure comparability between periods. These included mean daily worker and brood mortality, per-capita grooming rates averaged across three observation videos (30 min, 90 min, and one day post-introduction), mean daily egg-laying rate, recruitment expressed as the mean percentage of the colony present at the two food sources, and total food consumption corrected for evaporation.
Trade-off between resistance-related and tolerance-related strategies
For each experimental colony, we computed overall scores reflecting its investment in resistance- or tolerance-related strategies, calculated separately for control and infection periods. The resistance score was derived from four proxies: Δ allo-grooming events per capita, Δ extensive-grooming events per capita, Δ self-grooming events per capita, and necrophoresis speed (the latter applicable only for colonies under removable cadaver condition). For each proxy, the 14 sister colonies were ranked and assigned points from 1 (lowest value) to 14 (highest value). Resistance scores were obtained by summing points across all four resistance-related proxies and then normalized them between 0 and 1, where 1 represented the maximum possible score (14 points × 4 proxies = 56 points) and 0 the minimum (1 point × 4 proxies = 4 points). In the fixed cadaver condition, where necrophoresis could not be measured, the maximum possible score was adjusted to 42 points. The same procedure was applied to tolerance-related strategies, using two recruitment proxies (worker mobilization to food sources and food consumption) and two brood demographic proxies (egg-laying rate and the inverse of brood mortality rate). Tolerance scores were calculated by summing the ranked points across these proxies. To test for a trade-off between tolerance and resistance strategies, we examined the correlation between colonies’ tolerance and resistance scores using Pearson correlation tests (function cor.test, package stats).