Dataset and R code for: Banded mongooses discriminate relatedness and MHC diversity in unfamiliar conspecifics

This repository includes R code and data necessary to reproduce manuscript results and figures. The code has been written by Jamie Winternitz (jcwinternitz-at-gmail.com) and Hazel Nichols (H.J.Nichols-at-Swansea.ac.uk).<b>GENERAL INFORMATION</b>1. Paper citationSchubert N, Stober C, Born M, Mwanguhya F, Businge R, Kyambulima S, Mwesige K, Cant MA, Nichols HJ, Winternitz JC. 2025. Banded mongooses discriminate relatedness and MHC diversity in unfamiliar conspecifics. Behavioral Ecology.2. Brief abstractOlfactory cues play a vital role in mammalian social communication, conveying fitness-relevant information such as genetic quality and relatedness. Kin recognition through scent can help avoid inbreeding and guide nepotistic behaviors, enhancing fitness. In banded mongooses, synchronized breeding disrupts familiarity-based kin recognition, potentially increasing reliance on phenotype matching, where individuals compare genetically determined odors to assess similarity. We tested whether banded mongooses use odors to assess genetic diversity and relatedness based on (1) major histocompatibility complex (MHC) genotypes and (2) neutral microsatellite loci. Results showed individuals responded differently to odors from unfamiliar conspecifics based on MHC diversity and relatedness. Specifically, less MHC-diverse and less related individuals attracted more interest, suggesting odor cues are used to evaluate intruder or competitor threat levels. Neutral genetic diversity did not affect odor responses and was not correlated with MHC diversity, indicating responses to MHC diversity are independent of overall genetic diversity. No effect of MHC similarity was observed, possibly due to sample size limitations. Our findings suggest MHC diversity may signal genetic quality, while other genomic regions might contribute to assessing relatedness. These results provide a foundation for further research into the role of MHC and other genes in social communication in species where phenotype matching offers adaptive benefits.3. OriginatorsJamie Winternitz and Hazel Nichols4. Contact informationJamie Winternitz (jcwinternitz-at-gmail.com) and Hazel Nichols (H.J.Nichols-at-Swansea.ac.uk)5. Date of data collectionMay to July 20226. Geographic location of data collectionQueen Elizabeth National Park in Uganda (0°12’S, 27°54E’)7. Information about funding sources that supported the collection and curation of the dataThis work was supported by the German Research Foundation/Deutsche Forschungsgemeinschaft (DFG) (Project number 416495992 to JW; as part of the SFB TRR 212 (NC³) – Project numbers 316099922 and 396780709), by the Alexander von Humboldt Foundation (Research Fellowship to HJN), and by the Leverhulme Trust (International Fellowship, grant reference: IAF-2018-006 to HJN).<b>ACCESS INFORMATION</b>1. Licenses/restrictions placed on the dataCC BY 4.02. Data derived from other sourcesNot applicable<b>DATA FILES AND VARIABLES</b>1.<i> Full_scent_presentation_dataset.csv</i>Dataset used for scent presentation analyses.OdourID: Identifier of the individual providing the odor sample (odor donor).RecipID: Identifier of the individual receiving the odor presentation (odor recipient).Type: Presentation type category (i.e., unfamiliar).Control: Indicator variable denoting whether the trial was a control presentation (C = control, E = experimental).VideoN: Unique video identifier associated with each trial.Date: Date of the odor presentation (month/day/year).Trial: Unique trial identifier.Contact: Duration (s) the recipient spent in physical contact with the presented odor during the trial.Sniffing: Duration (s) the recipient spent sniffing the odor presentation during the trial.Rolling: Duration (s) the recipient spent rolling over the odor presentation during the trial.Licking: Duration (s) recipient spent liking the odor presentation during the trial.Duration: Total duration (s) until the recipient resumed normal foraging behavior after the presentation.Marking: Total number of scent marks deposited during the trial (could either be urine, feces or AGS markings).Overmark: Number of scent marks deposited directly on top of the odor mark (could either be urine, feces or AGS markings).Vicinity: Number of scent marks deposited within 30 cm of the odor mark (but not directly over it).Anal_gland_secretion: Indicator variable for deposited mark using anal gland secretion (AGS) (1 = yes, 0 = no).Urine: Indicator variable for deposited mark using urine (1 = yes, 0 = no).Faeces: Indicator variable for deposited mark using feces (1 = yes, 0 = no).Notes: Observational notes recorded during or after the trial.RecipPack: Social pack identity of the recipient.Sex: Sex of the recipient (0 = male, 1 = female).OdourPack: Social pack identity of the odor donor.Odoursx: Sex of the odor donor (0 = male, 1 = female).similarity: Pairwise genetic similarity between donor and recipient, based on MHC alleles.relatedness: Pairwise genetic relatedness between donor and recipient, based on microsatellite data.sMLH: Individual standardized multilocus heterozygosity calculated from microsatellite alleles.similarity.ST: Pairwise genetic similarity between donor and recipient, based on MHC supertypes.distinct.alleles: Number of distinct MHC alleles carried by the donor.distinct.supertypes: Number of distinct MHC supertypes carried by the donor.unique.reads: Number of unique long reads obtained for the donor from long-read sequencing.2. <i>sample_sets.csv</i>Table listing sampled individuals and the corresponding data available for that individual.sample_ID: Unique identifier of the individual contributing a sample to the datasetRecipID: Indicates whether the individual was an odor presentation recipient (1 = yes, 0 = no)OdourID: Indicates whether the individual was an odor donor (1 = yes, 0 = no)LRS_samples: Indicates whether the individual provided a gDNA sample for long-read sequencing (1 = yes, 0 = no)Geno_LRS_MHC: Indicates whether the individual has MHC genotypes from long-read sequencing (1 = yes, 0 = no)Geno_LRS: Indicates whether the individual has genotypes from long-read sequencing (1 = yes, 0 = no)Geno_LRS_unique_reads: Number of unique long reads obtained for the sample<br><b>CODE SCRIPTS AND WORKFLOW</b>1. <i>summary_of_datasets.R</i>R code that summarizes the structure of the different datasets used for the analyses.2. <i>r_code.R</i>R code that runs the analyses and produces the tables and figures in the manuscript and in the Supplementary Materials.<br><b>SOFTWARE VERSIONS</b>R version 4.4.0loaded packages:tidyverse 2.0.0afex 1.4-1lme4 1.1-37lmerTest 3.1-3ggeffects 2.3.0ggplot2 4.0.0broom 1.0.8purr 1.1.0ggpubr 0.6.1For questions or reproduction of results, please refer to the manuscript or contact the authors.