Clint Schwab
Research Subjects
Clint Schwab has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Francesco Tiezzi
Department of Animal Science, North Carolina State University, Raleigh, USA
7 shared publications - Christian Maltecca
Department of Animal Science, North Carolina State University
6 shared publications - Justin Fix
Acuity Ag Solutions, LLC, Carlyle, USA
4 shared publications - Caleb Shull
The Maschhoff’s, LLC, Carlyle, USA
3 shared publications - Nathan P. McNulty
Matatu Inc., St. Louis, USA
3 shared publications - Constantino Schillebeeckx
Matatu Inc., St. Louis, USA
2 shared publications
Latest Publications
- Open Access
- 7 June 2022
- Published by Springer Nature in Genetics Selection Evolution
- Vol. 54 (1) , 1-16
Genotyping and phenotyping strategies for genetic improvement of meat quality and carcass composition in swine
Background: Meat quality and composition traits have become valuable in modern pork production; however, genetic improvement has been slow due to high phenotyping costs. Combining genomic information with multi-trait indirect selection based on cheaper indicator traits is an alternative for continued cost-effective genetic improvement. Methods: Data from an ongoing breeding program were used in this study. Phenotypic and genomic information was collected on three-way crossbred and purebred Duroc animals belonging to 28 half-sib families. We applied different methods to assess the value of using purebred and crossbred information (both genomic and phenotypic) to predict expensive-to-record traits measured on crossbred individuals. Estimation of multi-trait variance components set the basis for comparing the different scenarios, together with a fourfold cross-validation approach to validate the phenotyping schemes under four genotyping strategies. Results: The benefit of including genomic information for multi-trait prediction depended on the breeding goal trait, the indicator traits included, and the source of genomic information. While some traits benefitted significantly from genotyping crossbreds (e.g., loin intramuscular fat content, backfat depth, and belly weight), multi-trait prediction was advantageous for some traits even in the absence of genomic information (e.g., loin muscle weight, subjective color, and subjective firmness). Conclusions: Our results show the value of using different sources of phenotypic and genomic information. For most of the traits studied, including crossbred genomic information was more beneficial than performing multi-trait prediction. Thus, we recommend including crossbred individuals in the reference population when these are phenotyped for the breeding objective.
- 7 June 2022
- Published by Springer Nature in Genetics Selection Evolution
- Vol. 54 (1) , 1-16
Open AccessGenotyping and phenotyping strategies for genetic improvement of meat quality and carcass composition in swine
Background: Meat quality and composition traits have become valuable in modern pork production; however, genetic improvement has been slow due to high phenotyping costs. Combining genomic information with multi-trait indirect selection based on cheaper indicator traits is an alternative for continued cost-effective genetic improvement. Methods: Data from an ongoing breeding program were used in this study. Phenotypic and genomic information was collected on three-way crossbred and purebred Duroc animals belonging to 28 half-sib families. We applied different methods to assess the value of using purebred and crossbred information (both genomic and phenotypic) to predict expensive-to-record traits measured on crossbred individuals. Estimation of multi-trait variance components set the basis for comparing the different scenarios, together with a fourfold cross-validation approach to validate the phenotyping schemes under four genotyping strategies. Results: The benefit of including genomic information for multi-trait prediction depended on the breeding goal trait, the indicator traits included, and the source of genomic information. While some traits benefitted significantly from genotyping crossbreds (e.g., loin intramuscular fat content, backfat depth, and belly weight), multi-trait prediction was advantageous for some traits even in the absence of genomic information (e.g., loin muscle weight, subjective color, and subjective firmness). Conclusions: Our results show the value of using different sources of phenotypic and genomic information. For most of the traits studied, including crossbred genomic information was more beneficial than performing multi-trait prediction. Thus, we recommend including crossbred individuals in the reference population when these are phenotyped for the breeding objective.
- Open Access
- 28 August 2021
- Published by Springer Nature in Animal Microbiome
- Vol. 3 (1) , 1-20
Microbial composition differs between production systems and is associated with growth performance and carcass quality in pigs
Background: The role of the microbiome in livestock production has been highlighted in recent research. Currently, little is known about the microbiome's impact across different systems of production in swine, particularly between selection nucleus and commercial populations. In this paper, we investigated fecal microbial composition in nucleus versus commercial systems at different time points. Results: We identified microbial OTUs associated with growth and carcass composition in each of the two populations, as well as the subset common to both. The two systems were represented by individuals with sizeable microbial diversity at weaning. At later times microbial composition varied between commercial and nucleus, with species of the genus Lactobacillus more prominent in the nucleus population. In the commercial populations, OTUs of the genera Lactobacillus and Peptococcus were associated with an increase in both growth rate and fatness. In the nucleus population, members of the genus Succinivibrio were negatively correlated with all growth and carcass traits, while OTUs of the genus Roseburia had a positive association with growth parameters. Lactobacillus and Peptococcus OTUs showed consistent effects for fat deposition and daily gain in both nucleus and commercial populations. Similarly, OTUs of the Blautia genus were positively associated with daily gain and fat deposition. In contrast, an increase in the abundance of the Bacteroides genus was negatively associated with growth performance parameters. Conclusions: The current study provides a first characterization of microbial communities' value throughout the pork production systems. It also provides information for incorporating microbial composition into the selection process in the quest for affordable and sustainable protein production in swine.
- 28 August 2021
- Published by Springer Nature in Animal Microbiome
- Vol. 3 (1) , 1-20
Open AccessMicrobial composition differs between production systems and is associated with growth performance and carcass quality in pigs
Background: The role of the microbiome in livestock production has been highlighted in recent research. Currently, little is known about the microbiome's impact across different systems of production in swine, particularly between selection nucleus and commercial populations. In this paper, we investigated fecal microbial composition in nucleus versus commercial systems at different time points. Results: We identified microbial OTUs associated with growth and carcass composition in each of the two populations, as well as the subset common to both. The two systems were represented by individuals with sizeable microbial diversity at weaning. At later times microbial composition varied between commercial and nucleus, with species of the genus Lactobacillus more prominent in the nucleus population. In the commercial populations, OTUs of the genera Lactobacillus and Peptococcus were associated with an increase in both growth rate and fatness. In the nucleus population, members of the genus Succinivibrio were negatively correlated with all growth and carcass traits, while OTUs of the genus Roseburia had a positive association with growth parameters. Lactobacillus and Peptococcus OTUs showed consistent effects for fat deposition and daily gain in both nucleus and commercial populations. Similarly, OTUs of the Blautia genus were positively associated with daily gain and fat deposition. In contrast, an increase in the abundance of the Bacteroides genus was negatively associated with growth performance parameters. Conclusions: The current study provides a first characterization of microbial communities' value throughout the pork production systems. It also provides information for incorporating microbial composition into the selection process in the quest for affordable and sustainable protein production in swine.
- Open Access
- 4 February 2021
- Published by Frontiers Media SA in Frontiers in Genetics
- Vol. 11
Genetic Parameters for Tolerance to Heat Stress in Crossbred Swine Carcass Traits
Data for loin and backfat depth, as well as carcass growth of 126,051 three-way crossbred pigs raised between 2015 and 2019, were combined with climate records of air temperature, relative humidity, and temperature–humidity index. Environmental covariates with the largest impact on the studied traits were incorporated in a random regression model that also included genomic information. Genetic control of tolerance to heat stress and the presence of genotype by environment interaction were detected. Its magnitude was more substantial for loin depth and carcass growth, but all the traits studied showed a different impact of heat stress and different magnitude of genotype by environment interaction. For backfat depth, heritability was larger under comfortable conditions (no heat stress), as compared to heat stress conditions. Genetic correlations between extreme values of environmental conditions were lower (∼0.5 to negative) for growth and loin depth. Based on the solutions obtained from the model, sires were ranked on their breeding value for general performance and tolerance to heat stress. Antagonism between overall performance and tolerance to heat stress was moderate. Still, the models tested can provide valuable information to identify genetic material that is resilient and can perform equally when environmental conditions change. Overall, the results obtained from this study suggest the existence of genotype by environment interaction for carcass traits, as a possible genetic contributor to heat tolerance in swine.
- 4 February 2021
- Published by Frontiers Media SA in Frontiers in Genetics
- Vol. 11
Open AccessGenetic Parameters for Tolerance to Heat Stress in Crossbred Swine Carcass Traits
Data for loin and backfat depth, as well as carcass growth of 126,051 three-way crossbred pigs raised between 2015 and 2019, were combined with climate records of air temperature, relative humidity, and temperature–humidity index. Environmental covariates with the largest impact on the studied traits were incorporated in a random regression model that also included genomic information. Genetic control of tolerance to heat stress and the presence of genotype by environment interaction were detected. Its magnitude was more substantial for loin depth and carcass growth, but all the traits studied showed a different impact of heat stress and different magnitude of genotype by environment interaction. For backfat depth, heritability was larger under comfortable conditions (no heat stress), as compared to heat stress conditions. Genetic correlations between extreme values of environmental conditions were lower (∼0.5 to negative) for growth and loin depth. Based on the solutions obtained from the model, sires were ranked on their breeding value for general performance and tolerance to heat stress. Antagonism between overall performance and tolerance to heat stress was moderate. Still, the models tested can provide valuable information to identify genetic material that is resilient and can perform equally when environmental conditions change. Overall, the results obtained from this study suggest the existence of genotype by environment interaction for carcass traits, as a possible genetic contributor to heat tolerance in swine.
- Open Access
- 29 July 2020
- Published by Springer Nature in Genetics Selection Evolution
- Vol. 52 (1) , 1-13
Modeling host-microbiome interactions for the prediction of meat quality and carcass composition traits in swine
Background: The objectives of this study were to evaluate genomic and microbial predictions of phenotypes for meat quality and carcass traits in swine, and to evaluate the contribution of host-microbiome interactions to the prediction. Data were collected from Duroc-sired three-way crossbred individuals (n = 1123) that were genotyped with a 60 k SNP chip. Phenotypic information and fecal 16S rRNA microbial sequences at three stages of growth (Wean, Mid-test, and Off-test) were available for all these individuals. We used fourfold cross-validation with animals grouped based on sire relatedness. Five models with three sets of predictors (full, informatively reduced, and randomly reduced) were evaluated. ‘Full’ included information from all genetic markers and all operational taxonomic units (OTU), while ‘informatively reduced’ and ‘randomly reduced’ represented a reduced number of markers and OTU based on significance preselection and random sampling, respectively. The baseline model included the fixed effects of dam line, sex and contemporary group and the random effect of pen. The other four models were constructed by including only genomic information, only microbiome information, both genomic and microbiome information, and microbiome and genomic information and their interaction. Results: Inclusion of microbiome information increased predictive ability of phenotype for most traits, in particular when microbiome information collected at a later growth stage was used. Inclusion of microbiome information resulted in higher accuracies and lower mean squared errors for fat-related traits (fat depth, belly weight, intramuscular fat and subjective marbling), objective color measures (Minolta a*, Minolta b* and Minolta L*) and carcass daily gain. Informative selection of markers increased predictive ability but decreasing the number of informatively reduced OTU did not improve model performance. The proportion of variation explained by the host-genome-by-microbiome interaction was highest for fat depth (~ 20% at Mid-test and Off-test) and shearing force (~ 20% consistently at Wean, Mid-test and Off-test), although the inclusion of the interaction term did not increase the accuracy of predictions significantly. Conclusions: This study provides novel insight on the use of microbiome information for the phenotypic prediction of meat quality and carcass traits in swine. Inclusion of microbiome information in the model improved predictive ability of phenotypes for fat deposition and color traits whereas including a genome-by-microbiome term did not improve prediction accuracy significantly.
- 29 July 2020
- Published by Springer Nature in Genetics Selection Evolution
- Vol. 52 (1) , 1-13
Open AccessModeling host-microbiome interactions for the prediction of meat quality and carcass composition traits in swine
Background: The objectives of this study were to evaluate genomic and microbial predictions of phenotypes for meat quality and carcass traits in swine, and to evaluate the contribution of host-microbiome interactions to the prediction. Data were collected from Duroc-sired three-way crossbred individuals (n = 1123) that were genotyped with a 60 k SNP chip. Phenotypic information and fecal 16S rRNA microbial sequences at three stages of growth (Wean, Mid-test, and Off-test) were available for all these individuals. We used fourfold cross-validation with animals grouped based on sire relatedness. Five models with three sets of predictors (full, informatively reduced, and randomly reduced) were evaluated. ‘Full’ included information from all genetic markers and all operational taxonomic units (OTU), while ‘informatively reduced’ and ‘randomly reduced’ represented a reduced number of markers and OTU based on significance preselection and random sampling, respectively. The baseline model included the fixed effects of dam line, sex and contemporary group and the random effect of pen. The other four models were constructed by including only genomic information, only microbiome information, both genomic and microbiome information, and microbiome and genomic information and their interaction. Results: Inclusion of microbiome information increased predictive ability of phenotype for most traits, in particular when microbiome information collected at a later growth stage was used. Inclusion of microbiome information resulted in higher accuracies and lower mean squared errors for fat-related traits (fat depth, belly weight, intramuscular fat and subjective marbling), objective color measures (Minolta a*, Minolta b* and Minolta L*) and carcass daily gain. Informative selection of markers increased predictive ability but decreasing the number of informatively reduced OTU did not improve model performance. The proportion of variation explained by the host-genome-by-microbiome interaction was highest for fat depth (~ 20% at Mid-test and Off-test) and shearing force (~ 20% consistently at Wean, Mid-test and Off-test), although the inclusion of the interaction term did not increase the accuracy of predictions significantly. Conclusions: This study provides novel insight on the use of microbiome information for the phenotypic prediction of meat quality and carcass traits in swine. Inclusion of microbiome information in the model improved predictive ability of phenotypes for fat deposition and color traits whereas including a genome-by-microbiome term did not improve prediction accuracy significantly.
- Open Access
Microbiability of meat quality and carcass composition traits in swine
The impact of gut microbiome composition was investigated at different stages of production (Wean, Mid-test, and Off-test) on meat quality and carcass composition traits of 1,123 three-way-crossbred pigs. Data were analyzed using linear mixed models which included the fixed effects of dam line, contemporary group and gender as well as the random effects of pen, animal and microbiome information at different stages. The contribution of the microbiome to all traits was prominent although it varied over time, increasing from weaning to Off-test for most traits. Microbiability estimates of carcass composition traits were greater compared to meat quality traits. Adding microbiome information did not affect the estimates of genomic heritability of meat quality traits but affected the estimates of carcass composition traits. High microbial correlations were found among different traits, particularly with traits related to fat deposition with decrease in genomic correlation up to 20% for loin weight and belly weight. Decrease in genomic heritabilities and genomic correlations with the inclusion of microbiome information suggested that genomic correlation was partially contributed by genetic similarity of microbiome composition.
Open AccessMicrobiability of meat quality and carcass composition traits in swine
The impact of gut microbiome composition was investigated at different stages of production (Wean, Mid-test, and Off-test) on meat quality and carcass composition traits of 1,123 three-way-crossbred pigs. Data were analyzed using linear mixed models which included the fixed effects of dam line, contemporary group and gender as well as the random effects of pen, animal and microbiome information at different stages. The contribution of the microbiome to all traits was prominent although it varied over time, increasing from weaning to Off-test for most traits. Microbiability estimates of carcass composition traits were greater compared to meat quality traits. Adding microbiome information did not affect the estimates of genomic heritability of meat quality traits but affected the estimates of carcass composition traits. High microbial correlations were found among different traits, particularly with traits related to fat deposition with decrease in genomic correlation up to 20% for loin weight and belly weight. Decrease in genomic heritabilities and genomic correlations with the inclusion of microbiome information suggested that genomic correlation was partially contributed by genetic similarity of microbiome composition.
- Open Access
Predicting Growth and Carcass Traits in Swine Using Metagenomic Data and Machine Learning Algorithms
In this paper, we evaluated the power of metagenome measures taken at three time points over the growth test period (weaning, 15 and 22 weeks) to foretell growth and carcass traits in 1039 individuals of a line of crossbred pigs. We measured prediction accuracy as the correlation between actual and predicted phenotypes in a five-fold cross-validation setting. Phenotypic traits measured included live weight measures and carcass composition obtained during the trial as well as at slaughter. We employed a null model excluding microbiome information as a baseline to assess the increase in prediction accuracy stemming from the inclusion of operational taxonomic units (OTU) as predictors. We further contrasted performance of models from the Bayesian alphabet (Bayesian Lasso) as well machine learning approaches (Random Forest and Gradient Boosting) and semi-parametric kernel models (Reproducing Kernel Hilbert space). In most cases, prediction accuracy increased significantly with the inclusion of microbiome data. Accuracy was more substantial with the inclusion of metagenomic information taken at week 15 and 22, with values ranging from approximately 0.30 for loin traits to more than 0.50 for back-fat. Conversely, microbiome composition at weaning resulted in most cases in marginal gains of prediction accuracy, suggesting that later measures might be more useful to include in predictive models. Model choice affected predictions marginally with no clear winner for any model/trait/time point. We, therefore, suggest average prediction across models as a robust strategy in fitting metagenomic information. In conclusion, microbiome composition can effectively be used as a predictor of growth and composition traits, particularly for fatness traits. The inclusion of OTU predictors could potentially be used to promote fast growth of individuals while limiting fat accumulation. Early microbiome measures might not be good predictors of growth and OTU information might be best collected at later life stages. Future research should focus on the inclusion of both microbiome as well as host genome information in predictions, as well as the interaction between the two. Furthermore, the influence of microbiome on feed efficiency as well as carcass and meat quality should be investigated.
Open AccessPredicting Growth and Carcass Traits in Swine Using Metagenomic Data and Machine Learning Algorithms
In this paper, we evaluated the power of metagenome measures taken at three time points over the growth test period (weaning, 15 and 22 weeks) to foretell growth and carcass traits in 1039 individuals of a line of crossbred pigs. We measured prediction accuracy as the correlation between actual and predicted phenotypes in a five-fold cross-validation setting. Phenotypic traits measured included live weight measures and carcass composition obtained during the trial as well as at slaughter. We employed a null model excluding microbiome information as a baseline to assess the increase in prediction accuracy stemming from the inclusion of operational taxonomic units (OTU) as predictors. We further contrasted performance of models from the Bayesian alphabet (Bayesian Lasso) as well machine learning approaches (Random Forest and Gradient Boosting) and semi-parametric kernel models (Reproducing Kernel Hilbert space). In most cases, prediction accuracy increased significantly with the inclusion of microbiome data. Accuracy was more substantial with the inclusion of metagenomic information taken at week 15 and 22, with values ranging from approximately 0.30 for loin traits to more than 0.50 for back-fat. Conversely, microbiome composition at weaning resulted in most cases in marginal gains of prediction accuracy, suggesting that later measures might be more useful to include in predictive models. Model choice affected predictions marginally with no clear winner for any model/trait/time point. We, therefore, suggest average prediction across models as a robust strategy in fitting metagenomic information. In conclusion, microbiome composition can effectively be used as a predictor of growth and composition traits, particularly for fatness traits. The inclusion of OTU predictors could potentially be used to promote fast growth of individuals while limiting fat accumulation. Early microbiome measures might not be good predictors of growth and OTU information might be best collected at later life stages. Future research should focus on the inclusion of both microbiome as well as host genome information in predictions, as well as the interaction between the two. Furthermore, the influence of microbiome on feed efficiency as well as carcass and meat quality should be investigated.
- Open Access
- 4 January 2018
- Published by Springer Nature in Microbiome
- Vol. 6 (1) , 1-15
Host contributes to longitudinal diversity of fecal microbiota in swine selected for lean growth
In pigs, gut bacteria have been shown to play important roles in nutritional, physiological, and immunological processes in the host. However, the contribution of their metagenomes or part of them, which are normally reflected by fragments of 16S rRNA-encoding genes, has yet to be fully investigated. Fecal samples, collected from a population of crossbred pigs at three time points, including weaning, week 15 post weaning (hereafter “week 15”), and end-of-feeding test (hereafter “off-test”), were used to evaluate changes in the composition of the fecal microbiome of each animal over time. This study used 1205, 1295, and 1283 samples collected at weaning, week 15, and off-test, respectively. There were 1039 animals that had samples collected at all three time points and also had phenotypic records on back fat thickness (BF) and average daily body weight gain (ADG). Firmicutes and Bacteroidetes were the most abundant phyla at all three time points. The most abundant genera at all three time points included Clostridium, Escherichia, Bacteroides, Prevotella, Ruminococcus, Fusobacterium, Campylobacter, Eubacterium, and Lactobacillus. Two enterotypes were identified at each time point. However, only enterotypes at week 15 and off-test were significantly associated with BF. We report herein two novel findings: (i) alpha diversity and operational taxonomic unit (OTU) richness were moderately heritable at week 15, h2 of 0.15 ± 0.06 to 0.16 ± 0.07 and 0.23 ± 0.09 to 0.26 ± 0.08, respectively, as well as at off-test, h2 of 0.20 ± 0.09 to 0.33 ± 0.10 and 0.17 ± 0.08 to 0.24 ± 0.08, respectively, whereas very low heritability estimates for both measures were detected at weaning; and (ii) alpha diversity at week 15 had strong and negative genetic correlations with BF, − 0.53 ± 0.23 to − 0.45 ± 0.25, as well as with ADG, − 0.53 ± 0.32 to − 0.53 ± 0.29. These results are important for efforts to genetically improve the domesticated pig because they suggest fecal microbiota diversity can be used as an indicator trait to improve traits that are expensive to measure.
- 4 January 2018
- Published by Springer Nature in Microbiome
- Vol. 6 (1) , 1-15
Open AccessHost contributes to longitudinal diversity of fecal microbiota in swine selected for lean growth
In pigs, gut bacteria have been shown to play important roles in nutritional, physiological, and immunological processes in the host. However, the contribution of their metagenomes or part of them, which are normally reflected by fragments of 16S rRNA-encoding genes, has yet to be fully investigated. Fecal samples, collected from a population of crossbred pigs at three time points, including weaning, week 15 post weaning (hereafter “week 15”), and end-of-feeding test (hereafter “off-test”), were used to evaluate changes in the composition of the fecal microbiome of each animal over time. This study used 1205, 1295, and 1283 samples collected at weaning, week 15, and off-test, respectively. There were 1039 animals that had samples collected at all three time points and also had phenotypic records on back fat thickness (BF) and average daily body weight gain (ADG). Firmicutes and Bacteroidetes were the most abundant phyla at all three time points. The most abundant genera at all three time points included Clostridium, Escherichia, Bacteroides, Prevotella, Ruminococcus, Fusobacterium, Campylobacter, Eubacterium, and Lactobacillus. Two enterotypes were identified at each time point. However, only enterotypes at week 15 and off-test were significantly associated with BF. We report herein two novel findings: (i) alpha diversity and operational taxonomic unit (OTU) richness were moderately heritable at week 15, h2 of 0.15 ± 0.06 to 0.16 ± 0.07 and 0.23 ± 0.09 to 0.26 ± 0.08, respectively, as well as at off-test, h2 of 0.20 ± 0.09 to 0.33 ± 0.10 and 0.17 ± 0.08 to 0.24 ± 0.08, respectively, whereas very low heritability estimates for both measures were detected at weaning; and (ii) alpha diversity at week 15 had strong and negative genetic correlations with BF, − 0.53 ± 0.23 to − 0.45 ± 0.25, as well as with ADG, − 0.53 ± 0.32 to − 0.53 ± 0.29. These results are important for efforts to genetically improve the domesticated pig because they suggest fecal microbiota diversity can be used as an indicator trait to improve traits that are expensive to measure.
- Open Access
- 1 January 2013
- Published by Springer Nature in BMC Genetics
- Vol. 14 (1) , 8
Methods of tagSNP selection and other variables affecting imputation accuracy in swine
Genotype imputation is a cost efficient alternative to use of high density genotypes for implementing genomic selection. The objective of this study was to investigate variables affecting imputation accuracy from low density tagSNP (average distance between tagSNP from 100kb to 1Mb) sets in swine, selected using LD information, physical location, or accuracy for genotype imputation. We compared results of imputation accuracy based on several sets of low density tagSNP of varying densities and selected using three different methods. In addition, we assessed the effect of varying size and composition of the reference panel of haplotypes used for imputation.
- 1 January 2013
- Published by Springer Nature in BMC Genetics
- Vol. 14 (1) , 8
Open AccessMethods of tagSNP selection and other variables affecting imputation accuracy in swine
Genotype imputation is a cost efficient alternative to use of high density genotypes for implementing genomic selection. The objective of this study was to investigate variables affecting imputation accuracy from low density tagSNP (average distance between tagSNP from 100kb to 1Mb) sets in swine, selected using LD information, physical location, or accuracy for genotype imputation. We compared results of imputation accuracy based on several sets of low density tagSNP of varying densities and selected using three different methods. In addition, we assessed the effect of varying size and composition of the reference panel of haplotypes used for imputation.