Opportunities and limits of combining microbiome and genome data for complex trait prediction

Abstract

Background Analysis and prediction of complex traits using microbiome data combined with host genomic information is a topic of utmost interest. However, numerous questions remain to be answered: how useful can the microbiome be for complex trait prediction? Are estimates of microbiability reliable? Can the underlying biological links between the host’s genome, microbiome, and phenome be recovered? Methods Here, we address these issues by (i) developing a novel simulation strategy that uses real microbiome and genotype data as inputs, and (ii) using variance-component approaches (Bayesian Reproducing Kernel Hilbert Space (RKHS) and Bayesian variable selection methods (Bayes C)) to quantify the proportion of phenotypic variance explained by the genome and the microbiome. The proposed simulation approach can mimic genetic links between the microbiome and genotype data by a permutation procedure that retains the distributional properties of the data. Results Using real genotype and rumen microbiota abundances from dairy cattle, simulation results suggest that microbiome data can significantly improve the accuracy of phenotype predictions, regardless of whether some microbiota abundances are under direct genetic control by the host or not. This improvement depends logically on the microbiome being stable over time. Overall, random-effects linear methods appear robust for variance components estimation, in spite of the typically highly leptokurtic distribution of microbiota abundances. The predictive performance of Bayes C was higher but more sensitive to the number of causative effects than RKHS. Accuracy with Bayes C depended, in part, on the number of microorganisms’ taxa that influence the phenotype. Conclusions While we conclude that, overall, genome-microbiome-links can be characterized using variance component estimates, we are less optimistic about the possibility of identifying the causative host genetic effects that affect microbiota abundances, which would require much larger sample sizes than are typically available for genome-microbiome-phenome studies. The R code to replicate the analyses is in https://github.com/miguelperezenciso/simubiome.

Genetic relationships between efficiency traits and gut microbiota traits in growing pigs being fed with a conventional or a high-fiber diet

“…The abundances of some genera were significantly correlated at the genetic level with feed efficiency traits, suggesting opportunities to use microbiota derived traits as proxies of feed efficiency in breeding. Following these results, some authors proposed to use the microbiota information to improve feed efficiency traits ( Weishaar et al, 2020 ; Christensen et al, 2021 ; Pérez-Enciso et al, 2021 ).…”

Gut microbiota and host genetics contribute to the phenotypic variation of digestive and feed efficiency traits in growing pigs fed a conventional and a high fiber diet

“…To confirm this, the predictive ability of the Micro+Gen model to predict phenotypes should be estimated and compared with that of the Micro and Gen models. Using simulated phenotypes in cattle, Pérez-Enciso et al [ 38 ] showed that including both microbiota and genome data in the model could increase the predictive accuracy by 50%. In addition, such comparisons could confirm if a model that fits jointly microbiota and genomic effects can help better predict digestive and feed efficiency phenotypes than models fitting each effect separately.…”

“…Microbiality estimates may be affected by the underlying model used for the distribution of the microbial effects, or by the estimation framework. Numerous methods exist to estimate , such as the Bayes C hypotheses of mixtures of distributions with different variances [ 38 ], MBLUP [ 18 ], and Bayesian regression [ 12 , 15 , 40 , 41 ]. Pérez-Enciso et al [ 38 ] compared estimates obtained with Bayes C and Bayesian RKHS.…”

“…Numerous methods exist to estimate , such as the Bayes C hypotheses of mixtures of distributions with different variances [ 38 ], MBLUP [ 18 ], and Bayesian regression [ 12 , 15 , 40 , 41 ]. Pérez-Enciso et al [ 38 ] compared estimates obtained with Bayes C and Bayesian RKHS. With the Bayes C method, they observed that the estimates can be biased, especially if is higher than 0.25, but for close to 0, they were well estimated.…”

Correction to: Opportunities and limits of combining microbiome and genome data for complex trait prediction

“…Following publication of the original article [ 1 ], the authors noticed the following two errors: Figures 1 and 2 are swapped, but the legends and figure numbers are correct, that is, the legend indicated for Figure 1 is in the right place but corresponds to Figure 2, and vice versa; and In the Funding section, the statement: MPE is funded by Ministry of Science and Innovation-State Research Agency (AEI) Grant PID2019-108829RB-I00 is incomplete. The correct paragraph should be: MPE is funded by the Ministry of Science and Innovation-State Research Agency (MICIN/AEI/10.13039/501100011033) Grant PID2019-108829RB-I00.…”