Food Reconstruction Using Isotopic Transferred Signals (FRUITS)

A Bayesian Model for Diet Reconstruction

Datos Bibliográficos

ID23340532
AutoresRicardo Fernandes (0000-0003-2258-3262, Christian-Albrechts-Universität zu Kiel, autor de correspondencia), Andrew R Millard (0000-0002-8290-7428, Durham University), Marek Brabec (0000-0001-6367-5791, Czech Academy of Sciences, Institute of Computer Science), Marie-Josée Nadeau, Marie‐josée Nadeau (0000-0001-9616-5632, Christian-Albrechts-Universität zu Kiel), Pieter M Grootes (0000-0003-4265-3168, Christian-Albrechts-Universität zu Kiel), Pieter Grootes
EditoresLuca Bondioli
Año2014
Volumen9
Número2
Páginase87436
Fecha de publicación2014-02-13
Peer ReviewedSí
Open AccessSí
TipoARTICLE
RevistaPLoS ONE (JOURNAL)
Identificadores de la revistaISSN: 1932-6203 • E-ISSN: 1932-6203
EditorialPublic Library of Science (PLoS) (PUBLISHER)
DOI10.1371/journal.pone.0087436
PMID24551057
PMCIDPMC3923756
OpenAlexW2031468096
IdiomaEN
Citas recibidas194
Referencias citadas32

Human and animal diet reconstruction studies that rely on tissue chemical signatures aim at providing estimates on the relative intake of potential food groups. However, several sources of uncertainty need to be considered when handling data. Bayesian mixing models provide a natural platform to handle diverse sources of uncertainty while allowing the user to contribute with prior expert information. The Bayesian mixing model FRUITS (Food Reconstruction Using Isotopic Transferred Signals) was developed for use in diet reconstruction studies. FRUITS incorporates the capability to account for dietary routing, that is, the contribution of different food fractions (e.g. macronutrients) towards a dietary proxy signal measured in the consumer. FRUITS also provides relatively straightforward means for the introduction of prior information on the relative dietary contributions of food groups or food fractions. This type of prior may originate, for instance, from physiological or metabolic studies. FRUITS performance was tested using simulated data and data from a published controlled animal feeding experiment. The feeding experiment data was selected to exemplify the application of the novel capabilities incorporated into FRUITS but also to illustrate some of the aspects that need to be considered when handling data within diet reconstruction studies. FRUITS accurately predicted dietary intakes, and more precise estimates were obtained for dietary scenarios in which expert prior information was included. FRUITS represents a useful tool to achieve accurate and precise food intake estimates in diet reconstruction studies within different scientific fields (e.g. ecology, forensics, archaeology, and dietary physiology).

Bayesian probability · Biology · Environmental health · Food group · Archaeology and ancient environmental studies · Artificial Intelligence · Computer Science · Food Science · Isotope Analysis in Ecology · Meat and Animal Product Quality · Medicine