PT  - JOURNAL ARTICLE
AU  - Stefanie Navaratnam
AU  - Julie Baker Phillips
TI  - Breaking the Scale: Allometric scaling analysis in Carnivoran families
AID  - 10.1101/2021.04.30.442221
DP  - 2021 Jan 01
TA  - bioRxiv
PG  - 2021.04.30.442221
4099  - http://biorxiv.org/content/early/2021/05/01/2021.04.30.442221.short
4100  - http://biorxiv.org/content/early/2021/05/01/2021.04.30.442221.full
AB  - The analysis of scaling relationships, allometric scaling, has a long history of importance for modelling and predicting biological phenomena. Individual organisms are not truly independent, and as a result phylogenetic corrections are necessary to increase the accuracy of scaling relationships. The relationships between body mass and gestation length have not been previously reported at the Family level, as it was previously thought species had insufficient time to diverge evolutionarily leaving phylogenetic corrections unnecessary. Using a Carnivora supertree, we perform a phylogenetically generalised least squares (PGLS) analysis using life history information largely from the Pantheria dataset. Our results suggest that allometric relationships are maintained in four families: Canidae, Felidae, Herpestidae and Otariidae. Conversely, several evolutionary mechanisms may contribute to the lack of a significant scaling parameter in other families, such as diverse reproductive strategies or positive selection for genes affecting adiposity. In addition, low sample sizes or the inclusion of paternal body masses could alter the presence of significant scaling. Our results suggests that PGLS analyses are informative at the family level, and the absence of scaling can provide insight to understanding of the evolutionary mechanisms that work on the family taxonomic level or below.CCS CONCEPTS • Applied computing → Molecular evolution; • Computing methodologies → Modeling methodologies.ACM Reference Format Stefanie Navaratnam and Julie Baker Phillips. 2021. Breaking the Scale: Allometric scaling analysis in Carnivoran families. In BCB: ACM Conference on Bioinformatics, Computational Biology, and Health Informatics, August 01–04, 2021, Virtual due to COVID-19. ACM, New York, NY, USA, 7 pages. https://doi.org/10.1145/1122445.1122456Competing Interest StatementThe authors have declared no competing interest.