Using Precision Phenotyping to Inform de novo Domestication

One-sentence summary: An update on the use of precision phenotyping to assess the potential of lesser cultivated species as candidates for de novo domestication or similar development for future agriculture.

379 ASAS-EAAP Talk: Precision Phenotyping using Infrared Spectroscopy to Improve the Quality of Animal Products

There is an ever-growing interest in research oriented towards the improvement of quality of animal products. In this context, one major operational bottleneck is the possibility to collect quality indicators over the meat and dairy chains and for selective breeding purposes. The use of near-infrared (NIR) and the Fourier-transformed infrared (FTIR) spectroscopy techniques have been proven to be powerful precision phenotyping tools for high-throughput meat and milk quality assessment. Such technologies allow scoring large number of animals and/or derived-products for novel (predicted) phenotypes and indicator traits to set-up potential new payment systems and boost the genetic improvement. One important step in the use of NIR and FTIR tools is the definition of the “gold standard” as the infrared-based predictions could act only as indicators traits. Indeed, the definition of a robust calibration set, the assessment of repeatability and reproducibility of the reference (i.e., gold standard) as well as the detection of random and systematic errors are crucial steps. Once the reference phenotype has been defined, different statistical methodologies could be applied to infrared spectra data. For instance, the partial least squares regression (PLS) is a multivariate regression method commonly used to build up prediction models using NIR and FTIR spectra data. However, the implementation of advanced statistical approaches, such as Bayesian approaches and machine learning methods, might allow us to achieve more robust and accurate predictions. In this talk, we will describe and discuss some of the challenges and potentials of NIR and FTIR tools for large-scale precision phenotyping. Some examples include the use of NIR and Visible-NIR (Vis-NIR) for assessing meat quality parameters (also using portable instruments able to collect spectra directly from the muscle surface at the slaughterhouse) and the use of FTIR for predicting several traits related to fine milk composition and technological traits in dairy cattle.

Artificial Intelligence in Cardiac Imaging

Machine learning (ML), a subset of artificial intelligence, is showing promising results in cardiology, especially in cardiac imaging. ML algorithms are allowing cardiologists to explore new opportunities and make discoveries not seen with conventional approaches. This offers new opportunities to enhance patient care and open new gateways in medical decision-making. This review highlights the role of ML in cardiac imaging for precision phenotyping and prognostication of cardiac disorders.