Breeding goals & design

The problem of using ratio traits when improving feed and methane efficiency

---

Abstract

Animal breeding programs aim to improve productivity while reducing input costs and resource use. As sustainability has become a central concern, feed intake and enteric methane emissions now feature more prominently in breeding objectives, supported by increasing availability of individual-animal measurements. With expanding phenotypic datasets for methane output in sheep and cattle, an important question is how to incorporate these traits most effectively in selection indices aimed at improving both resource efficiency and environmental outcomes. A seemingly appealing strategy is to use ratio traits, such as production per unit of feed intake or methane emitted per unit of product, as direct indicators of efficiency. However, ratio traits pose several conceptual and statistical problems. Including a ratio (x/y) in a breeding objective is mathematically equivalent to including its component traits with fixed relative weights determined solely by their population means (μy for x and -μx for y). These weights do not reflect true economic values or environmental priorities and therefore may misalign selection responses with industry goals. When ratio traits are used as selection criteria, the response to selection becomes driven largely by the genetic coefficient of variation (sa/μ) of each component trait, further distorting the intended direction of genetic change. Residual traits-such as residual feed intake (RFI) or residual methane-are sometimes proposed as alternatives to ratio traits. Using residual traits in breeding objectives or indices is statistically equivalent to including the component traits directly, with appropriate adjustments to genetic parameters. Conditional traits are, by definition, uncorrelated with the traits on which they are conditioned, but this also means that economic weights must be recalibrated to reflect the true biological and economic consequences of altering these traits. Defining residual traits at the phenotypic level is less desirable because it requires that both component phenotypes be measured on the same animals. Instead, EBVs for residual traits can be generated from EBVs of the underlying traits using their estimated variances and covariances. A more robust strategy is to include methane production directly in a selection index and optimise index weights according to an explicit breeding objective that balances profitability with reduced methane output at the industry level. The choice of reference point is critical, because reducing methane per animal does not necessarily reduce methane per unit of product or per industry-wide output. Genetic correlations among methane, feed intake and production traits will therefore determine the achievable improvement in methane efficiency. Given current carbon prices, a desired-gains index may be required, as economic signals alone are unlikely to sufficiently incentivise breeding for reduced methane emissions.

Keywords: 2026

How to Cite:

van der Werf, J., (2026) “The problem of using ratio traits when improving feed and methane efficiency”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2289213. doi: https://doi.org/10.31274/wcgalp.24298

Rights: 1