Genetic basis of kin discrimination in Bacillus subtilisMršol, Manca (Avtor) Štefanič, Polonca (Mentor) da Rocha , Ulisses Nunes (Komentor) kin discriminationBacillus subtilismachine learningaverage nucleotide identityKin discrimination (KD) is the capacity to distinguish individuals based on their (phylo)genetic relation, a behavior seen in organisms like the Gram-positive soil bacterium Bacillus subtilis. In this study, a detailed investigation of 40 B. subtilis strains was conducted to identify and characterize unique features specific to its KD groups. The groups were determined by observing how strains interact with each other during swarming on agar plates. The analysis of 40 complete of B. subtilis genomes, employing genome annotation tools such as Prokka, PGAP, PredicTF, and BioAutoML along with manual curation of the comQXP locus, contributed to the creation of a new binary dataset. This dataset led to the discovery of specific genetic markers unique to each kin group, thereby uncovering the diverse genetic composition of B. subtilis. The identified genetic elements were predominantly related to cell wall synthesis, stress response, and antibiotic/antimicrobial properties. Additionally, the study utilized ML algorithms in combination with core genome alignement and Average Nucleotide Identity (ANI) for classifying kin groups and developed a pipeline integrating both methods. Currently, ANI surpasses the ML approach in performance. However, to truly identify unique markers we would need a bigger dataset with more genomes and further laboratory research to verify the accuracy of both methods and the identified markers.[M. Mršol]20242024-09-25 07:16:21Magistrsko delo/naloga162523sl