The topic of feed efficiency is under discussion in the scientific community and several studies pointed out that lactation stage has to be accounted for when estimates of feed efficiency are carried out, especially for genetic ranking of animals and their performances, as highlighted by Li et al. (2017). Other researchers applied a latin square design to test dietary effects across lactation (Ipharraguerre et al. 2002) but this approach cannot be followed out of experimental conditions and particularly does not allow, nowadays, to valorize precision livestock farm data to get phenotypic information from individual animals at farm level. 

The current manuscript by Fischer, et al. (2022a) describes an experimental trial in which cows were first fed a high starch diet-low fibre then switched over to a low starch diet-high fibre and individually monitored over time. Data were analyzed with the objective to investigate effects within diets and across diets. Since all cows went through the same sequence at the same time it was not possible to completely separate the confounding effect of lactation stage and diet as stated by the authors. However, this manuscript adds methodological discussions and opens research questions especially to the matter of repeatability and reproducibility of feed efficiency of individual animals over the lactation stage. These variables are fundamental to evaluate nutritional traits and phenotypic performances of dairy cows at farm level, as highlighted by a paper of the same first author (Fischer, et al. 2022b) dealing to reproducibility and repeatability with a similar approach. My opinion is that this manuscript gives the opportunity to enlarge the scientific discussions on the calculation of repeatability and reproducibility of feed efficiency of individual animals over time. In particular, as in this study, specific mathematical approaches need to be carried out with the final goal to analyze and valorize precision livestock farm data for cow phenotyping and to propose new methods of feed efficiency evaluations. It also needs complete databases carried out under experimental conditions. In fact it has to be considered that this manuscript makes available to the scientific community all the data and the R code developed for data analysis giving the opportunity to replicate the calculations and propose new advancements in the feed efficiency evaluations of dairy cows.

References 

Fischer A, Gasnier P, Faverdin P (2022a) Feed efficiency of lactating Holstein cows was not as repeatable across diets as within diet over subsequent lactation stages. bioRxiv, 2021.02.10.430560, ver. 3 peer-reviewed and recommended by Peer Community in Animal Science. https://doi.org/10.1101/2021.02.10.430560

Fischer A, Dai X, Kalscheur KF (2022b) Feed efficiency of lactating Holstein cows is repeatable within diet but less reproducible when changing dietary starch and forage concentrations. animal, 16, 100599. https://doi.org/10.1016/J.ANIMAL.2022.100599

Ipharraguerre IR, Ipharraguerre RR, Clark JH (2002) Performance of Lactating Dairy Cows Fed Varying Amounts of Soyhulls as a Replacement for Corn Grain. Journal of Dairy Science, 85, 2905–2912. https://doi.org/10.3168/JDS.S0022-0302(02)74378-6

Li B, Berglund B, Fikse WF, Lassen J, Lidauer MH, Mäntysaari P, Løvendahl P (2017) Neglect of lactation stage leads to naive assessment of residual feed intake in dairy cattle. Journal of Dairy Science, 100, 9076–9084. https://doi.org/10.3168/JDS.2017-12775

Using sequence data in livestock genomics has often been regarded as a solution to revolutionize livestock breeding (Meuwissen & Goddard, 2010). The main expected benefits were to enhance the accuracy of breeding values, achieve better persistence of the accuracy over generations, and enable across populations or breed predictions (Hickey, 2013). Despite the promised benefits, whole-genome sequencing has not yet been implemented in livestock breeding programs, replacing SNP arrays for routine evaluation.

In this work, Johnsson (2023) thoroughly reviewed the literature regarding the implications of whole-genome sequencing and functional genomics for livestock breeding practice. The author discusses the potential applications and reasons for difficulties in their implementation. The author speculates that the main challenge for making using the sequence data profitable is to overcome the problem of the small dimensionality of the genetic data and proposes three potential ways to achieve this goal. The first approach is better modeling of genomic segments, the second inclusion of undetected genetic variation, and the third use of functional genomic information.  

The paper presents an original and interesting perspective on the current status of the use of sequence data in livestock breeding programs and perspectives for the future. 

References

Hickey,J.M.,2013.Sequencing millions of animals for genomic selection 2.0. Journal of Animal Breeding and Genetics 130:331–332. https://doi.org/10.1111/jbg.12054 

Johnsson, M., 2023. The big challenge for livestock genomics is to make sequence data pay. arXiv, 2302.01140, ver. 4 peer-reviewed and recommended by Peer Community in Animal Science. https://doi.org/10.48550/arXiv.2302.01140 

Meuwissen, T., Goddard, M.,2010. Accurate prediction of genetic values for complex traits by whole-genome resequencing. Genetics 185:623–631. https://doi.org/10.1534/genetics.110.116590 

Farm animals differ in their ability to respond to the many environmental challenges they face. Such challenges include infectious diseases, metabolic diseases resulting from inadequate coverage of dietary needs, as well as the diverse consequences of climate change. Various concepts exist to characterise the responses of animals to different types of challenges. This article by Friggens et al. (2022) focuses on resilience, providing a conceptual definition and proposing a method to quantify resilience in dairy cows.

The first part of the paper provides a definition of resilience and highlights its differences and relations with the related concepts of robustness, and, to a lesser extent, resistance and tolerance. In essence, resilience is the ability of an animal to bounce back quickly after a challenge of limited duration. On the other hand, robustness is the ability of an animal to cope with conditions that are overall unfavourable. From these conceptual and intuitive definitions, there are several difficulties precluding the design of concrete methods to measure resilience. First, there is some degree of overlap between the concepts of resilience, robustness, resistance and tolerance. Secondly, resilience is a multidimensional concept whereby resilience to a given perturbation does not imply resilience to other types of perturbation, e.g. resilience to a challenge by a specific pathogen does not imply resilience to a nutritional challenge. A further difficulty in the measure of resilience is the fact that different animals may be exposed to challenges that are different in nature and in number. The authors argue that although resilience cannot be measured directly (it should be seen as a latent construct), it is possible to quantify it indirectly through its consequences.

In the second part of the paper, the authors propose a method to quantify resilience of individual dairy cows. The method is based on the premise that resilient animals should be kept longer in their herd than non-resilient animals. The main criterion in the evaluation is therefore the ability of cows to re-calve. Each cow that is calving receives a certain number of points, to  which, in each lactation, bonus points are added for higher milk production and penalty points are removed for each insemination after the first one, for each disease event and for each day of calving interval above some herd specific value. Therefore, cows have a resilience score in each lactation. They also have a lifetime resilience score obtained by summing the scores for all the lactations, that gets bigger as the cow has more calves, and that also takes the age at first calving into account. In a previous study, Adriaens et al. (2020) showed that higher resilience scores were associated with fewer drops in milk yield and more stable activity dynamics.

Starting from theoretical considerations on the notion of resilience, this paper describes a concrete method to quantify animal-level resilience on farm. Such quantification will be useful for breeding and culling decisions. Finally, the general framework to design resilience measures that is presented will be useful to researchers working on the quantification of farm animal resilience using new methods and data sources.

References

Adriaens I, Friggens NC, Ouweltjes W, Scott H, Aernouts B and Statham J 2020. Productive life span and resilience rank can be predicted from on-farm first-parity sensor time series but not using a common equation,  across farms. Journal of Dairy Science 103, 7155-7171.https://doi.org/10.3168/jds.2019-17826

Friggens, N.C. , Adriaens, I., Boré, R., Cozzi, G., Jurquet, J., Kamphuis, C., Leiber, F., Lora, I., Sakowski, T., Statham, J., De Haas, Y. (2022). Resilience: reference measures based on longer-term consequences are needed to unlock the potential of precision livestock farming technologies for quantifying this trait. Zenodo, 5215797, ver. 5 peer-reviewed and recommended by Peer community in Animal Science. https://dx.doi.org/10.5281/zenodo.5215797

The adaptability of livestock to changing environments is based in particular on their genetic characteristics but also on the farming conditions to which they are subjected. However, this last point is poorly documented and little is known about its contribution to environmental challenges. The study by Quéméner and colleagues [1] addresses this question by assessing the effect of two hygiene conditions (good vs poor) on the distribution of cell populations present in adipose and muscle tissues of pigs divergently selected for feed efficiency [2].

The working hypothesis is that degraded housing conditions would be at the origin of an hyper stimulation of the immune system that can influence the homeostasis of adipose tissue and skeletal muscle and consequently modulate the cellular content of these tissues. Cellular compositions are thus interesting intermediate phenotypes for quantifying complex traits. The study uses pigs divergently selected for residual feed intake (RFI+ and RFI-) to assess whether there is a genetic effect associated with the observed phenotypes. 

The study characterized different stromal cell populations based on the expression of surface markers: CD45 to separate hematopoietic lineages and markers associated with the stem properties of mesenchymal cells: CD56, CD34, CD38 and CD140a. The authors observed that certain subpopulations are differentially enriched according to the hygiene condition (good vs poor) in adipose and skeletal tissue (CD45-CD56-) sometimes with an associated (genetic) lineage effect. This pioneering study validates a number of tools for characterizing cell subpopulations present in porcine adipose and muscle tissue. It confirms that housing conditions can have an effect on intermediate phenotypes such as intra-tissue cell populations. This pioneering work will pave the way to better understand the effects of livestock systems on tissue biology and animal phenotypes and to characterize the nature and function of progenitor cells present in muscle and adipose tissue.

[1] Quéméner A, Dessauge F, Perruchot MH, Le Floc’h N, Louveau I. 2022. The impact of housing conditions on porcine mesenchymal stromal/stem cell populations differ between adipose tissue and skeletal muscle. bioRxiv 2021.06.08.447546, ver. 3 peer-reviewed and recommended by Peer Community in Animal Science. https://doi.org/10.1101/2021.06.08.447546 

[2] Gilbert H, Bidanel J-P, Gruand J, Caritez J-C, Billon Y, Guillouet P, Lagant H, Noblet J, Sellier P. 2007. Genetic parameters for residual feed intake in growing pigs, with emphasis on genetic relationships with carcass and meat quality traits. Journal of Animal Science 85:3182–3188. https://doi.org/10.2527/jas.2006-590.

In this paper, Madouasse et al. (2021) present a generalisable Bayesian method for calculating the probability that a herd is free from disease, based on its prior disease status, and using data (herd status over time over a sufficient number of herds to inform the model) and reasonable prior estimates of the sensitivity and specificity of tests being used to determine animal infection status.  Where available, the modelling approach can also include relevant additional risk factors. 

By bringing all these factors together, it allows for most countries to use the same analytical approach on their data, with differences across datasets expressed in terms of the uncertainty around the central estimates. 

Having a single methodology that generates both a central estimate of disease freedom, and uncertainty thus provides the opportunity (given typically available data) to compare the probability of freedom across different systems. This is relevant in terms of the context of trade (since international trade of livestock in many cases depends on disease freedom). It is also important when evaluating, for example, transnational burdens of disease - and with different regulations in place in different countries, this is invaluable and can be used, for example, to assess risks of zoonotic infection including for zoonotic infection emergence. In the BVD example provided, the point is made that, since regular testing would probably pick up infection rapidly, the addition of risk factors is most valuable where testing is infrequent. This emphasizes the advantages of direct incorporation of risk factors into a single modelling framework. 

From a technical point of view, the analysis compares two different packages for the Markov Chain Monte Carlo (MCMC) implementation necesary to run the model. They show that, while there are some slight systematic differences, the estimates provided by the two methods are similar to each other; as one method is approximate but substantially more stable and generally much more computationally efficient, this is an important outcome. Both implementations are freely available and  with relevant additional software made similarly available by the authors. This is extremely welcome and should encourage its general adoption across different countries. 

No single model can of course account for everything. In particular, the reliance on past data means that there is an implicit assumption common to all purely statistical methods that the underlying risks have not changed. Thus projections to altered circumstances (changing underlying risk factors or systematic changes in testing or test performance) cannot so easily be incorporated, since these factors are complicated by the dynamics of infection that lie outside the modelling approach. Of course the well known quote from George Box that "all models are wrong" applies here - the generality of approach, statistical robustness and open source philosophy adopted make this model very useful indeed.  

Madouasse A, Mercat M, van Roon A, Graham D, Guelbenzu M, Santman Berends I, van Schaik G, Nielen M, Frössling J, Ågren E, Humphry RW, Eze J, Gunn GJ, Henry MK, Gethmann J, More SJ, Toft N, Fourichon C (2021) A modelling framework for the prediction of the herd-level probability of infection from longitudinal data. bioRxiv, 2020.07.10.197426, ver. 6 peer-reviewed and recommended by PCI Animal Science. https://doi.org/10.1101/2020.07.10.197426