Removal of alleles by genome editing – RAGE against the deleterious load

Background In this paper, we simulate deleterious load in an animal breeding program, and compare the efficiency of genome editing and selection for decreasing load. Deleterious variants can be identified by bioinformatics screening methods that use sequence conservation and biological prior information about protein function. Once deleterious variants have been identified, how can they be used in breeding? Results We simulated a closed animal breeding population subject to both natural selection against deleterious load and artificial selection for a quantitative trait representing the breeding goal. Deleterious load was polygenic and due to either codominant or recessive variants. We compared strategies for removal of deleterious alleles by genome editing (RAGE) to selection against carriers. Each strategy varied in how animals and variants were prioritized for editing or selection. Conclusions Genome editing of deleterious alleles reduces deleterious load, but requires simultaneous editing of multiple deleterious variants in the same sire to be effective when deleterious variants are recessive. In the short term, selection against carriers is a possible alternative to genome editing when variants are recessive. The dominance of deleterious variants affects both the efficiency of genome editing and selection against carriers, and which variant prioritization strategy is the most efficient. Our results suggest that in the future, there is the potential to use RAGE against deleterious load in animal breeding.

compared strategies for removal of deleterious alleles by genome editing (RAGE) to 23 selection against carriers. Each strategy varied in how animals and variants were prioritized 24 for editing or selection. 25

Conclusions 26
Genome editing of deleterious alleles reduces deleterious load, but requires simultaneous 27 editing of multiple deleterious variants in the same sire to be effective when deleterious 28 variants are recessive. In the short term, selection against carriers is a possible alternative to 29 genome editing when variants are recessive. The dominance of deleterious variants affects 30 both the efficiency of genome editing and selection against carriers, and which variant 31 prioritization strategy is the most efficient. Our results suggest that in the future, there is the 32 potential to use RAGE against deleterious load in animal breeding. 33 against carriers is the strategy of choice for removing monogenic recessive deleterious 66 variants from animal breeding populations [5,33]. Analogously, one could select against 67 deleterious load by avoiding selection candidates with high deleterious load. 68

69
The aim of this paper was to compare the efficiency of genome editing and selection against 70 carriers for decreasing deleterious load in an animal breeding program. We simulated 71 polygenic deleterious load subject to natural selection in a simulation of a closed animal 72 breeding population artificially selected for a quantitative performance trait representing the 73 breeding goal. We compared removal of alleles by genome editing (RAGE) to selection 74 against carriers using genotypes at deleterious variants. We compared strategies for 75 prioritizing variants for editing and individuals for selection based on deleterious allele and 76 genotype frequencies. Our results showed that RAGE reduces deleterious load, but requires 77 simultaneous editing of multiple deleterious variants in the same sire to be effective when 78 deleterious variants are recessive. In the short term, selection against carriers is a possible 79 alternative to genome editing when variants are recessive, but in the future, RAGE against 80 the deleterious load has great potential in animal breeding. 81 82

83
We used simulations to compare genome editing and selection against carriers using 84 genotypes at deleterious variants. We simulated artificial selection for a quantitative trait 85 variants simultaneously, we modelled a quantitative breeding goal trait and a fitness trait. 112

113
The breeding goal trait was a polygenic quantitative trait with additive effects. We randomly 114 assigned 10 000 segregating sites (1000 per chromosome) as quantitative trait variants for the 115 breeding goal trait with additive effects drawn from a normal distribution. 116 117 Fitness was a polygenic multiplicative trait that represented probability of survival prior to 118 artificial selection. We randomly assigned 10 000 segregating sites as fitness variants (again 119 1000 per chromosome), choosing variants that with allele frequencies below 0.01 for 120 codominant variants and 0.1 for recessive variants. The fitness variants were chosen 121 independently of the quantitative trait variants. The deleterious effect size was expressed as a 122 selection coefficient s against the mutant allele, going from 0 (no deleterious effect) to 1 (a 123 lethal allele). The fitness of each genotype was 1 for the homozygous wildtype, 1 -h s for the 124 heterozygote, and 1 -s for the mutant homozygote, where h is a dominance coefficient. 125 Dominance coefficients were either 0 for recessive variants or 0.5 codominant variants. We 126 assumed multiplicative effects, so that the fitness of an individual was the product of the 127 contribution of each fitness variant. The effect sizes were drawn from a mixture of three 128 uniform distributions with one third of variants being small (0 < s < 10 -4 ), one third 129 intermediate (10 -4 < s < 0.1), and one third large (0.1 < s < 1). These proportions were chosen 130 based on the estimated distribution of deleterious effects in humans [13]. 131 132 Deleterious mutations occurred randomly during burn-in and historical breeding with a per 133 locus mutation rate of 10 -4 , to give a deleterious mutation rate of 1 per individual and 134 genome. This is a conservative estimate for the deleterious mutation rate in mammals. No 135 back-mutation was allowed, meaning that only wild type alleles could mutate. Quantitative 136 trait variants for the breeding goal trait did not mutate, except during the initial coalescent 137 simulation to create ancestral haplotypes. To simulate deleterious variant discovery, we selected a random fraction of the deleterious 152 variants that segregated at the end of historical breeding and assumed them to be discovered. 153 We used a discovery rate of 0.75 for the main scenarios, but also tested discovery rates of 0.5 154 and 1. To simulate imperfect detection of deleterious variants, we chose neutral segregating 155 variants as false positives at random. We added false positives so that the total number of 156 variants detected was equal to the number of segregating deleterious variants, and if 157 discovery rate was d, a fraction 1-d were false positives. These discovered variants were 158 allowed to be edited or used for selection against carriers subsequently. 159 During future breeding, we removed alleles by genome editing at discovered deleterious 163 variants in all the selected sires. For variants where a sire was not already homozygous 164 wildtype, we edited the genotype to homozygous wildtype, until a set number of variants had 165 been edited. We assumed that editing was accurate in that it always produced wild type 166 homozygotes, and had no deleterious off target effects. We edited 1, 5, or 20 variants per sire. 167 We only edited variants that were discovered and segregating in the population. 168

169
We used five different strategies for prioritizing variants for editing. These strategies were 170 based on information that would be available from genotyping the sires at discovered 171 deleterious variants, namely the deleterious allele and genotype frequencies. We assumed that 172 the deleterious effect size was unknown. For comparison, we also ran a baseline scenario without selection against carriers, starting 206 from the same initial populations after historical breeding.  This difference in the distribution of frequencies and effects translates to differences in the 256 efficiency of RAGE and selection against carriers. Figure 2 shows a comparison of genome 257 editing using the best performing variant prioritization strategy, and selection against carriers 258 using total deleterious load. When deleterious variants were codominant, the best performing 259 strategy was prioritizing low frequency variants for removal by editing, and selection against 260 carriers was inefficient. When deleterious variants were recessive, the best performing The efficiency of genome editing of deleterious variants was affected by the number of 281 variants edited per sire, and the strategy for prioritizing variants for editing. Figure 5 shows 282 and prioritizing low frequency, high frequency, or randomly chosen deleterious variants for 284 editing. Figure 6 shows trajectories of fitness during future breeding using variant 285 prioritization strategies devised for recessive variants: prioritizing variants with intermediate 286 frequency by applying an allele frequency threshold of 0.25, and editing variants based on 287 their deficit of homozygotes. In all these cases, the discovery rate was 0.75, meaning that 288 75% of segregating deleterious variants were discovered, and a 25% false positive rate. Because codominant deleterious alleles are expressed even when in a heterozygous state, they 364 are exposed to purifying selection. Therefore, the best variant prioritization strategy was to 365 prioritize low frequency variants for editing. coefficients for all variants. In real genomes, we expect that many more than 10 000 sites can 425 give rise to deleterious mutations, but since the number of segregating variants was not much 426 affected by the total number of fitness variants in the genome, this assumption appears to 427 have little impact on results. We simulated fitness as independent of the selected performance 428 trait. In real populations, we expect that fitness is to some extent already part of the breeding 429 goal in the form of survival, fecundity, and health traits. This means that it is possibly to 430 validate deleterious variants by phenotypic means, and including them in genomic selection 431 estimate, given that deleterious mutation rates for humans are often estimated to be higher 433 We found that genome editing of deleterious alleles reduced deleterious load, but that when 439 variants were recessive, simultaneous editing of multiple deleterious variants in the same sire 440 was needed for it to be competitive with selection against carriers. When accurate multiplex 441 genome editing becomes available, RAGE has the potential to improve fitness to levels that 442 are impossible by selection against carriers. This is a formidable undertaking, but a possible 443 long term goal. The long-term benefits of genome editing to remove deleterious variants over 444 selection against carriers include both the possibility of greater gains in fitness, and the ability 445 to improve fitness without sacrificing selection intensity for the breeding goal trait. 446

447
In the short term, selection against carriers is a possible alternative to genome editing. It is 448 ineffective against codominant variants, but when variants are recessive, it is more effective 449 at alleviating deleterious load than editing one variant per sire, but it is less effective than 450 multiplex editing. The cost of multiplex genome editing is unknown, but can assumed to be 451 high. Therefore, it appears that selection against carriers will remain superior for some time. 452 The downside of selection against carriers is that the number of sires available for selection is 453  One genomic feature that may play into prioritization of deleterious variants for genome 479 editing is recombination rate variation. In regions of low recombination, which in mammal 480 genomes occur for example in centromeric regions and on the sex chromosomes, selection 481 phenomenon may both lead to accumulation of deleterious variants and reduced selection for 483 beneficial variants that are located there. Therefore, it may also be beneficial to prioritize 484 variants that experience low recombination rate for genome editing [62,63]. 485

486
We have investigated removal of alleles by genome editing and selection against carriers to 487 alleviate the load of deleterious variants that segregate within a population. Genome editing 488 could also be used to remove deleterious alleles that are fixed in the population, and cannot 489 be removed by selection. Fixed deleterious variants could be detected by sequencing studies 490 that sample across populations and breeds that carry different sets of deleterious variants due 491 to chance events such as mutation, genetic drift, and founder effects.