Shape and object data analysis: Shape and object data analysis

Reconstructing past migratory behaviour of reindeer (Rangifer tarandus): Insights from geometric morphometric analysis of proximal phalanx morphology from extant caribou populations

Reindeer mobility patterns vary widely in modern ecosystems, notably between more open or more wooded environments. This renders the reconstruction of past reindeer mobility patterns challenging, while being at the same time key if archaeologists want to better understand the role that reindeer herds played in the subsistence and territorial organisation of Prehistoric hunter-gatherer societies. Studying the morphology associated with different habitats and mobility patterns can be a useful method for understanding past reindeer behaviour. To access paleoecological information, the relationship between locomotor anatomy and substrate type can be explored in modern animals and transposed to the past, as previous studies have proven that an animal´s environment affects bone morphology. In this study, 3D Geometric Morphometrics are used to explore the impact of extant reindeer habitat type and mobility pattern on phalanx morphology. Results obtained reflects on the potential archaeological application of such an approach for paleoecological reconstructions. Size and shape of phalanx vary significantly, yet complex to interpret in light of interplaying factors such as subspecies, sexual dimorphism and the influence of migration costs, snow cover and substrate type. If direct application to the archaeological record remains preliminary, this first study permits to highlight promising avenues for future research.

Bone adaptation: Safety factors and load predictability in shaping skeletal form

Much is known about skeletal adaptation in relation to the mechanical functions that bones serve. This includes how bone adapts to mechanical loading during an individual's lifetime as well as over evolutionary time. Although controlled loading in animal models allows us to observe short-term bone adaptation (epigenetic mechanobiology), examining an assemblage of extant vertebrate bones or a group of fossils' bony structures can reveal the combined effects of long-term trends in loading history and the effects of natural selection. In this survey we examine adaptations that take place over both time scales and highlight a few of the extraordinary insights first published by John Currey. First, we provide a historical perspective on bone adaptation control mechanisms, followed by a discussion of safety factors in bone. We then summarize examples of structural- and material-level adaptations and mechanotransduction, and analyze the relationship between these structural- and material-level adaptations observed in situations where loading modes are either predictable or unpredictable. We argue that load predictability is a major consideration for bone adaptation broadly across an evolutionary timescale, but that its importance can also be seen during ontogenetic growth trajectories, which are subject to natural selection as well. Furthermore, we suggest that bones with highly predictable load patterns demonstrate more precise design with lower safety factors, while bones that experience less predictable loads or those that are less capable of repair and adaptation are designed with a higher safety factor. Finally, exposure to rare loading events with high potential costs of failure leads to design of structures with very high safety factor compared to everyday loading experience. Understanding bone adaptations at the structural and material levels, which take place over an individual's lifetime or over evolutionary time has numerous applications in translational and clinical research to understand and treat musculoskeletal diseases, as well as to permit the furthering of human extraterrestrial exploration in environments with altered gravity.

A general framework for functional regression modelling

Researchers are increasingly interested in regression models for functional data. This article discusses a comprehensive framework for additive (mixed) models for functional responses and/or functional covariates based on the guiding principle of reframing functional regression in terms of corresponding models for scalar data, allowing the adaptation of a large body of existing methods for these novel tasks. The framework encompasses many existing as well as new models. It includes regression for ‘generalized’ functional data, mean regression, quantile regression as well as generalized additive models for location, shape and scale (GAMLSS) for functional data. It admits many flexible linear, smooth or interaction terms of scalar and functional covariates as well as (functional) random effects and allows flexible choices of bases—particularly splines and functional principal components—and corresponding penalties for each term. It covers functional data observed on common (dense) or curve-specific (sparse) grids. Penalized-likelihood-based and gradient-boosting-based inference for these models are implemented in R packages refund and FDboost , respectively. We also discuss identifiability and computational complexity for the functional regression models covered. A running example on a longitudinal multiple sclerosis imaging study serves to illustrate the flexibility and utility of the proposed model class. Reproducible code for this case study is made available online.