Decoding diets: a novel DNA-based approach for identifying cephalopods from beaks

Cephalopods play a key role in marine ecosystems but are difficult to study directly; hence, beaks recovered from predator stomachs are critical to further understand their distribution and ecology. Cephalopod species have traditionally been identified from beak morphology, a time-consuming method limited by degree of beak digestion, lack of specialist expertise, and limited coverage in reference collections. Here, we provide the first detailed methodology to identify cephalopod species through DNA analysis of their beaks. By extracting, amplifying and sequencing the cytochrome c oxidase I (COI) gene, we successfully annotated DNA sequences to species level in 50.5% of samples and 60% of beaks, obtained from predator stomachs and scientific nets. There was no statistically significant effect of beak section (untanned, tanned, rostrum tip), storage condition, or bleach treatment on assignment success (GLMM; all p > 0.05). Although formal statistical comparison was not possible because collection method and storage condition were confounded, success rates appeared to vary by sample origin, with lower success in samples from albatrosses (Thalassarche spp.) boluses than either stomachs of toothfish (Dissostichus spp.) or scientific nets (27.3%, 59.5% and 55.6%, respectively). Our results show that DNA analysis from cephalopod beaks can enable species-level identification, though expanding genetic databases with sequences from additional taxa and populations is crucial to improve accuracy. This approach enhances the recovery of ecological data from samples that were previously challenging to identify, supporting biodiversity monitoring, food-web research, and assessments of ecosystem change in rapidly shifting oceans.