Jung-Woo Choi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jung-Woo Choi has a total of 117 co-authors. Here are the top co-authors by number of shared publications:
- Eun-Seok Cho
Swine Division, National Institute of Animal Science, Rural Development Administration, Cheonan 31000, Korea
9 shared publications - Won-Hyong Chung
Department of Animal Science, College of Animal Life Sciences, Kangwon National University, Chuncheon 24341, Republic of Korea
6 shared publications Ha-Seung Seong
Swine Science Division
5 shared publications- Jin Soo Kim
Kangwon National University
5 shared publications - Joon-Hee Lee
Department of Animal Bioscience, College of Agriculture and Life Sciences, Gyeongsang National University, Jinju 52828, Republic of Korea
5 shared publications - Yong-Min Kim
Swine Science Division, National Institute of Animal Science, Rural Development Administration, Cheonan 31000, Republic of Korea
5 shared publications
Latest Publications
- Open Access
Development and Characterization of a Single Nucleotide Polymorphism Genotyping Panel for Duck Populations
Single nucleotide polymorphism (SNP) genotyping panels for domesticated ducks are currently lacking, despite their importance for genetic diversity studies and genomic selection. We analyzed whole-genome sequencing data from 74 ducks. To obtain a 40K SNP marker set, we performed SNP calling and applied minor allele frequency (MAF) and linkage disequilibrium (LD) criteria. SNPs from 33 genes associated with Quantitative Trait Loci (QTL) were manually selected and added to the panel. The performance of the resulting SNP panel was validated using genotyping data from 28 additional duck samples. We developed a 40K SNP genotyping panel for ducks. Applying criteria of call rate, MAF, and LD, 37,386 SNP markers were selected. After excluding SNPs near repeats and indels, and incorporating 591 SNPs from 33 duck QTL-associated genes, the final set consisted of 35,613 SNP markers. When evaluating the performance of the SNP panel, genotypes were successfully called for 35,209 out of 35,613 SNPs, achieving a call rate of 98.9%. After applying QC filters, 33,870 SNPs were confirmed to be usable. Our SNP panel design process successfully generated a dataset at a 40K scale without compromising the genetic distances of the whole genome. By minimizing variation at the SNP probe alignment sites, we achieved a high call rate. The panel demonstrated reasonable robustness and suitability for further genomic studies, making it a valuable tool for genetic improvement and trait selection in duck populations.
Open AccessDevelopment and Characterization of a Single Nucleotide Polymorphism Genotyping Panel for Duck Populations
Single nucleotide polymorphism (SNP) genotyping panels for domesticated ducks are currently lacking, despite their importance for genetic diversity studies and genomic selection. We analyzed whole-genome sequencing data from 74 ducks. To obtain a 40K SNP marker set, we performed SNP calling and applied minor allele frequency (MAF) and linkage disequilibrium (LD) criteria. SNPs from 33 genes associated with Quantitative Trait Loci (QTL) were manually selected and added to the panel. The performance of the resulting SNP panel was validated using genotyping data from 28 additional duck samples. We developed a 40K SNP genotyping panel for ducks. Applying criteria of call rate, MAF, and LD, 37,386 SNP markers were selected. After excluding SNPs near repeats and indels, and incorporating 591 SNPs from 33 duck QTL-associated genes, the final set consisted of 35,613 SNP markers. When evaluating the performance of the SNP panel, genotypes were successfully called for 35,209 out of 35,613 SNPs, achieving a call rate of 98.9%. After applying QC filters, 33,870 SNPs were confirmed to be usable. Our SNP panel design process successfully generated a dataset at a 40K scale without compromising the genetic distances of the whole genome. By minimizing variation at the SNP probe alignment sites, we achieved a high call rate. The panel demonstrated reasonable robustness and suitability for further genomic studies, making it a valuable tool for genetic improvement and trait selection in duck populations.
- Open Access
- 1 May 2026
- Published by Asian Australasian Association of Animal Production Societies in Animal Bioscience
- Vol. 39 (5)
Development and validation of a dual-purpose 60K single nucleotide polymorphism chip for simultaneous genotyping of chicken and duck
Objective: Chickens and ducks represent the most important poultry species within the livestock industry. However, the availability of single nucleotide polymorphism (SNP) chips for these species is limited, and development of separate chips for each species requires considerable expense. To address this limitation, we developed the first poultry SNP chip applicable to both species simultaneously. The 60K SNP chip comprises 30,816 SNPs for chickens and 35,209 SNPs for ducks. The performance of the chip was evaluated using non-pooled datasets (chicken only and duck only) and a pooled dataset combining chicken and duck DNA to verify that genotyping accuracy was maintained without crossspecies interference.Methods: DNA was extracted from 28 Korean native chickens and 28 Korean native ducks. Genotyping was performed using the Illumina platform, employing chicken-only, duck-only, and pooled chicken-duck SNP panels. Genotype accuracy and concordance were evaluated with PLINK. Sample and SNP quality were assessed in accordance with the Illumina genotyping protocol. Principal component analysis (PCA) was conducted to evaluate and compare the chicken-duck pooled dataset with the non-pooled datasets.Results: The mean genotype concordance of non-pooled (chicken only and duck only) datasets exceeded 99%, while concordance between non-pooled and chicken-duck pooled datasets surpassed 97.8%. All datasets satisfied quality thresholds for call rate, GenCall score, 10% GenCall score, and cluster separation. PCA revealed consistent clustering patterns, with no significant differences observed between non-pooled and pooled datasets.Conclusion: The first, at the best of our knowledge, poultry SNP chip incorporating chicken-duck pooled samples, has been successfully developed and validated. This chip provides reliable genotyping performance for both species and presents a cost-effective option for large-scale SNP chip development in the livestock industry.
- 1 May 2026
- Published by Asian Australasian Association of Animal Production Societies in Animal Bioscience
- Vol. 39 (5)
Open AccessDevelopment and validation of a dual-purpose 60K single nucleotide polymorphism chip for simultaneous genotyping of chicken and duck
Objective: Chickens and ducks represent the most important poultry species within the livestock industry. However, the availability of single nucleotide polymorphism (SNP) chips for these species is limited, and development of separate chips for each species requires considerable expense. To address this limitation, we developed the first poultry SNP chip applicable to both species simultaneously. The 60K SNP chip comprises 30,816 SNPs for chickens and 35,209 SNPs for ducks. The performance of the chip was evaluated using non-pooled datasets (chicken only and duck only) and a pooled dataset combining chicken and duck DNA to verify that genotyping accuracy was maintained without crossspecies interference.Methods: DNA was extracted from 28 Korean native chickens and 28 Korean native ducks. Genotyping was performed using the Illumina platform, employing chicken-only, duck-only, and pooled chicken-duck SNP panels. Genotype accuracy and concordance were evaluated with PLINK. Sample and SNP quality were assessed in accordance with the Illumina genotyping protocol. Principal component analysis (PCA) was conducted to evaluate and compare the chicken-duck pooled dataset with the non-pooled datasets.Results: The mean genotype concordance of non-pooled (chicken only and duck only) datasets exceeded 99%, while concordance between non-pooled and chicken-duck pooled datasets surpassed 97.8%. All datasets satisfied quality thresholds for call rate, GenCall score, 10% GenCall score, and cluster separation. PCA revealed consistent clustering patterns, with no significant differences observed between non-pooled and pooled datasets.Conclusion: The first, at the best of our knowledge, poultry SNP chip incorporating chicken-duck pooled samples, has been successfully developed and validated. This chip provides reliable genotyping performance for both species and presents a cost-effective option for large-scale SNP chip development in the livestock industry.
- Open Access
- 1 August 2025
- Published by Asian Australasian Association of Animal Production Societies in Animal Bioscience
- Vol. 38 (8) , 1594-1604
Selection signature analysis using whole genome resequencing data reveals candidate genes for white plumage color in Korean native ducks
Objective: Domestication alters the phenotypes of wild animals to meet human demands and leaves characteristic patterns in their genomes. Various selection signature analysis methods have been developed to identify these characteristic patterns left in the genome. The Korean native duck (KND) is one of the domesticated species in Korea. KND is categorized into two populations based on plumage color; colored KND (KNDC) and white KND (KNDW). To enhance the competitiveness of native ducks, it is necessary to establish a KNDW line. In this study, we conducted selection signature analysis to identify candidate genes associated with white plumage color in KNDs.Methods: We generated whole genome resequencing data from 22 KNDCs, 22KNDWs, and 10 Pekin ducks (PKDs). To detect distinct genomic regions between KND populations with different plumage colors, we analyzed three types of selection signature analysis: the fixation index (Fst), nucleotide diversity(π), and cross-population extended haplotype heterozygosity (XP-EHH).Results: Population structure analysis showed that although KNDC and KNDW are distinct from PKD, they form a single group sharing a common ancestor. The results of Fst and π analyses revealed that compared to KNDC, there were strong selection signals in the MITF gene in KNDW, with a 6,641 bp insertion in the intron 2 region. This variant is a transposable element insertion that causes white plumage in PKD. In addition, XP-EHH analysis identified DCT, KIT, TYR, and ADCY9 as major candidate genes associated with pigmentation in the KND population.Conclusion: White plumage in KNDW is caused by a transposable element insertion in the MITF gene. This finding improves our understanding of plumage color in KND and supports the establishment of KNDW breeding programs.
- 1 August 2025
- Published by Asian Australasian Association of Animal Production Societies in Animal Bioscience
- Vol. 38 (8) , 1594-1604
Open AccessSelection signature analysis using whole genome resequencing data reveals candidate genes for white plumage color in Korean native ducks
Objective: Domestication alters the phenotypes of wild animals to meet human demands and leaves characteristic patterns in their genomes. Various selection signature analysis methods have been developed to identify these characteristic patterns left in the genome. The Korean native duck (KND) is one of the domesticated species in Korea. KND is categorized into two populations based on plumage color; colored KND (KNDC) and white KND (KNDW). To enhance the competitiveness of native ducks, it is necessary to establish a KNDW line. In this study, we conducted selection signature analysis to identify candidate genes associated with white plumage color in KNDs.Methods: We generated whole genome resequencing data from 22 KNDCs, 22KNDWs, and 10 Pekin ducks (PKDs). To detect distinct genomic regions between KND populations with different plumage colors, we analyzed three types of selection signature analysis: the fixation index (Fst), nucleotide diversity(π), and cross-population extended haplotype heterozygosity (XP-EHH).Results: Population structure analysis showed that although KNDC and KNDW are distinct from PKD, they form a single group sharing a common ancestor. The results of Fst and π analyses revealed that compared to KNDC, there were strong selection signals in the MITF gene in KNDW, with a 6,641 bp insertion in the intron 2 region. This variant is a transposable element insertion that causes white plumage in PKD. In addition, XP-EHH analysis identified DCT, KIT, TYR, and ADCY9 as major candidate genes associated with pigmentation in the KND population.Conclusion: White plumage in KNDW is caused by a transposable element insertion in the MITF gene. This finding improves our understanding of plumage color in KND and supports the establishment of KNDW breeding programs.
- Open Access
Detection of Copy Number Variations in Woori-Heukdon Populations with the Illumina PorcineSNP60 Bead-Chip Array
This study investigated copy number variations (CNVs) in 2112 pigs from five populations: Korean Duroc (DUC), Korean Native Pig (KNP), and their crossbred offspring (F1, F2, and WRH). CNVs were detected using PennCNV and QuantiSNP, with CNVRuler identifying 698 CNV regions (CNVRs), covering 109 Mb (4.83%) of the porcine genome. Comparison with previous CNV studies on swine revealed CNVR overlap rates ranging from 31.12% (French Yorkshire) to 81.27% (Xiang), and 9.06% newly identified CNVRs. DUC showed the most CNVRs (n = 384), followed by WRH (n = 225). Meanwhile, F1 and F2 exhibited far fewer CNVRs (five and seven, respectively). Functional enrichment analysis highlighted various genes overlapping with the CNVRs, including 1236 genes in DUC and 572 genes in WRH, linked to biological processes. The quantitative trait loci (QTLs), overlapping with CNVRs, exhibited particular overlapping with traits such as average daily gain (4.24% of QTLs in DUC, 4.51% of QTLs in WRH). In contrast, KNP, F1, and F2 populations exhibited a higher frequency of CNVRs containing QTLs overlapped with drip loss. These findings indicate that WRH may inherit growth traits from DUC. This study provides a better understanding of CNVs in the pigs, which can potentially be used in improving genetic merits of pig populations.
Open AccessDetection of Copy Number Variations in Woori-Heukdon Populations with the Illumina PorcineSNP60 Bead-Chip Array
This study investigated copy number variations (CNVs) in 2112 pigs from five populations: Korean Duroc (DUC), Korean Native Pig (KNP), and their crossbred offspring (F1, F2, and WRH). CNVs were detected using PennCNV and QuantiSNP, with CNVRuler identifying 698 CNV regions (CNVRs), covering 109 Mb (4.83%) of the porcine genome. Comparison with previous CNV studies on swine revealed CNVR overlap rates ranging from 31.12% (French Yorkshire) to 81.27% (Xiang), and 9.06% newly identified CNVRs. DUC showed the most CNVRs (n = 384), followed by WRH (n = 225). Meanwhile, F1 and F2 exhibited far fewer CNVRs (five and seven, respectively). Functional enrichment analysis highlighted various genes overlapping with the CNVRs, including 1236 genes in DUC and 572 genes in WRH, linked to biological processes. The quantitative trait loci (QTLs), overlapping with CNVRs, exhibited particular overlapping with traits such as average daily gain (4.24% of QTLs in DUC, 4.51% of QTLs in WRH). In contrast, KNP, F1, and F2 populations exhibited a higher frequency of CNVRs containing QTLs overlapped with drip loss. These findings indicate that WRH may inherit growth traits from DUC. This study provides a better understanding of CNVs in the pigs, which can potentially be used in improving genetic merits of pig populations.
- Open Access
- 31 December 2024
- Published by The Korean Society of Poultry Science in Korean Journal of Poultry Science
- Vol. 51 (4) , 295-300
Genetic Diversity Analysis of Colored and White Feathered Native Duck Population
한국가금학회지(The Journal of Curriculum Evaluation)
- 31 December 2024
- Published by The Korean Society of Poultry Science in Korean Journal of Poultry Science
- Vol. 51 (4) , 295-300
Open AccessGenetic Diversity Analysis of Colored and White Feathered Native Duck Population
한국가금학회지(The Journal of Curriculum Evaluation)
- Open Access
- 31 December 2023
- Published by The Korean Society of Animal Reproduction and Biotechnology (KSARB) in Journal of Animal Reproduction and Biotechnology
- Vol. 38 (4) , 275-290
Comparative pluripotent characteristics of porcine induced pluripotent stem cells generated using different viral transduction systems
Sang-Ki Baek, In-Won Lee, Yeon-Ji Lee, Bo-Gyeong Seo, Jung-Woo Choi, Tae-Suk Kim, Cheol Hwangbo and Joon-Hee Lee. Journal of Animal Reproduction and Biotechnology 2023;38:275-90. https://doi.org/10.12750/JARB.38.4.275
- 31 December 2023
- Published by The Korean Society of Animal Reproduction and Biotechnology (KSARB) in Journal of Animal Reproduction and Biotechnology
- Vol. 38 (4) , 275-290
Open AccessComparative pluripotent characteristics of porcine induced pluripotent stem cells generated using different viral transduction systems
Sang-Ki Baek, In-Won Lee, Yeon-Ji Lee, Bo-Gyeong Seo, Jung-Woo Choi, Tae-Suk Kim, Cheol Hwangbo and Joon-Hee Lee. Journal of Animal Reproduction and Biotechnology 2023;38:275-90. https://doi.org/10.12750/JARB.38.4.275
- Open Access
- 31 March 2023
- Published by Korean Society of Animal Science and Technology in Journal of Animal Science and Technology
- Vol. 65 (2) , 401-411
Profiling of differentially expressed proteins between fresh and frozen-thawed Duroc boar semen using ProteinChip CM10
Journal of Animal Science and Technology (JAST) is a peer-reviewed, open access journal publishing original research, review articles and notes in all fields of animal science. Topics covered by the journal include: genetics and breeding, physiology, nutrition of monogastric animals, nutrition of ruminants, animal products (milk, meat, eggs and their by-products) and their processing, grasslands and roughages, livestock environment, animal biotechnology, animal behavior and welfare.
- 31 March 2023
- Published by Korean Society of Animal Science and Technology in Journal of Animal Science and Technology
- Vol. 65 (2) , 401-411
Open AccessProfiling of differentially expressed proteins between fresh and frozen-thawed Duroc boar semen using ProteinChip CM10
Journal of Animal Science and Technology (JAST) is a peer-reviewed, open access journal publishing original research, review articles and notes in all fields of animal science. Topics covered by the journal include: genetics and breeding, physiology, nutrition of monogastric animals, nutrition of ruminants, animal products (milk, meat, eggs and their by-products) and their processing, grasslands and roughages, livestock environment, animal biotechnology, animal behavior and welfare.
- Open Access
Identification of Copy Number Variations in Four Horse Breed Populations in South Korea
In this study, genome-wide CNVs were identified using a total of 469 horses from four horse populations (Jeju horses, Thoroughbreds, Jeju riding horses, and Hanla horses). We detected a total of 843 CNVRs throughout all autosomes: 281, 30, 301, and 310 CNVRs for Jeju horses, Thoroughbreds, Jeju riding horses, and Hanla horses, respectively. Of the total CNVRs, copy number losses were found to be the most abundant (48.99%), while gains and mixed CNVRs accounted for 41.04% and 9.96% of the total CNVRs, respectively. The length of the CNVRs ranged from 0.39 kb to 2.8 Mb, while approximately 7.2% of the reference horse genome assembly was covered by the total CNVRs. By comparing the CNVRs among the populations, we found a significant portion of the CNVRs (30.13%) overlapped; the highest number of shared CNVRs was between Hanla horses and Jeju riding horses. When compared with the horse CNVRs of previous studies, 26.8% of CNVRs were found to be uniquely detected in this study. The CNVRs were not randomly distributed throughout the genome; in particular, the Equus caballus autosome (ECA) 7 comprised the largest proportion of its genome (16.3%), while ECA 24 comprised the smallest (0.7%). Furthermore, functional analysis was applied to CNVRs that overlapped with genes (genic-CNVRs); these overlapping areas may be potentially associated with the olfactory pathway and nervous system. A racing performance QTL was detected in a CNVR of Thoroughbreds, Jeju riding horses, and Hanla horses, and the CNVR value was mixed for three breeds.
Open AccessIdentification of Copy Number Variations in Four Horse Breed Populations in South Korea
In this study, genome-wide CNVs were identified using a total of 469 horses from four horse populations (Jeju horses, Thoroughbreds, Jeju riding horses, and Hanla horses). We detected a total of 843 CNVRs throughout all autosomes: 281, 30, 301, and 310 CNVRs for Jeju horses, Thoroughbreds, Jeju riding horses, and Hanla horses, respectively. Of the total CNVRs, copy number losses were found to be the most abundant (48.99%), while gains and mixed CNVRs accounted for 41.04% and 9.96% of the total CNVRs, respectively. The length of the CNVRs ranged from 0.39 kb to 2.8 Mb, while approximately 7.2% of the reference horse genome assembly was covered by the total CNVRs. By comparing the CNVRs among the populations, we found a significant portion of the CNVRs (30.13%) overlapped; the highest number of shared CNVRs was between Hanla horses and Jeju riding horses. When compared with the horse CNVRs of previous studies, 26.8% of CNVRs were found to be uniquely detected in this study. The CNVRs were not randomly distributed throughout the genome; in particular, the Equus caballus autosome (ECA) 7 comprised the largest proportion of its genome (16.3%), while ECA 24 comprised the smallest (0.7%). Furthermore, functional analysis was applied to CNVRs that overlapped with genes (genic-CNVRs); these overlapping areas may be potentially associated with the olfactory pathway and nervous system. A racing performance QTL was detected in a CNVR of Thoroughbreds, Jeju riding horses, and Hanla horses, and the CNVR value was mixed for three breeds.
- Open Access
Functional Characterization of Endothelial Cells Differentiated from Porcine Epiblast Stem Cells
Endothelial cells (ECs), lining blood vessels’ lumen, play an essential role in regulating vascular functions. As multifunctional components of vascular structures, pluripotent stem cells (PSCs) are the promising source for potential therapeutic applications in various vascular diseases. Our laboratory has previously established an approach for differentiating porcine epiblast stem cells (pEpiSCs) into ECs, representing an alternative and potentially superior cell source. However, the condition of pEpiSCs-derived ECs growth has yet to be determined, and whether pEpiSCs differentiate into functional ECs remained unclear. Changes in morphology, proliferation and functional endothelial marker were assessed in pEpiSCs-derived ECs in vitro. pEpiSCs-derived ECs were subjected to magnetic-activated cell sorting (MACS) to collect CD-31+ of ECs. We found that sorted ECs showed the highest proliferation rate in differentiation media in primary culture and M199 media in the subculture. Next, sorted ECs were examined for their ability to act as typical vascular ECs through capillary-like structure formation assay, Dil-acetylated low-density lipoprotein (Dil-Ac-LDL) uptake, and three-dimensional spheroid sprouting. Consequently, pEpiSCs-derived ECs function as typical vascular ECs, indicating that pEpiSC-derived ECs might be used to develop cell therapeutics for vascular disease.
Open AccessFunctional Characterization of Endothelial Cells Differentiated from Porcine Epiblast Stem Cells
Endothelial cells (ECs), lining blood vessels’ lumen, play an essential role in regulating vascular functions. As multifunctional components of vascular structures, pluripotent stem cells (PSCs) are the promising source for potential therapeutic applications in various vascular diseases. Our laboratory has previously established an approach for differentiating porcine epiblast stem cells (pEpiSCs) into ECs, representing an alternative and potentially superior cell source. However, the condition of pEpiSCs-derived ECs growth has yet to be determined, and whether pEpiSCs differentiate into functional ECs remained unclear. Changes in morphology, proliferation and functional endothelial marker were assessed in pEpiSCs-derived ECs in vitro. pEpiSCs-derived ECs were subjected to magnetic-activated cell sorting (MACS) to collect CD-31+ of ECs. We found that sorted ECs showed the highest proliferation rate in differentiation media in primary culture and M199 media in the subculture. Next, sorted ECs were examined for their ability to act as typical vascular ECs through capillary-like structure formation assay, Dil-acetylated low-density lipoprotein (Dil-Ac-LDL) uptake, and three-dimensional spheroid sprouting. Consequently, pEpiSCs-derived ECs function as typical vascular ECs, indicating that pEpiSC-derived ECs might be used to develop cell therapeutics for vascular disease.
- Open Access
Use of Host Feeding Behavior and Gut Microbiome Data in Estimating Variance Components and Predicting Growth and Body Composition Traits in Swine
The purpose of this study was to investigate the use of feeding behavior in conjunction with gut microbiome sampled at two growth stages in predicting growth and body composition traits of finishing pigs. Six hundred and fifty-one purebred boars of three breeds: Duroc (DR), Landrace (LR), and Large White (LW), were studied. Feeding activities were recorded individually from 99 to 163 days of age. The 16S rRNA gene sequences were obtained from each pig at 123 ± 4 and 158 ± 4 days of age. When pigs reached market weight, body weight (BW), ultrasound backfat thickness (BF), ultrasound loin depth (LD), and ultrasound intramuscular fat (IMF) content were measured on live animals. Three models including feeding behavior (Model_FB), gut microbiota (Model_M), or both (Model_FB_M) as predictors, were investigated. Prediction accuracies were evaluated through cross-validation across genetic backgrounds using the leave-one-breed-out strategy and across rearing environments using the leave-one-room-out approach. The proportions of phenotypic variance of growth and body composition traits explained by feeding behavior ranged from 0.02 to 0.30, and from 0.20 to 0.52 when using gut microbiota composition. Overall prediction accuracy (averaged over traits and time points) of phenotypes was 0.24 and 0.33 for Model_FB, 0.27 and 0.19 for Model_M, and 0.40 and 0.35 for Model_FB_M for the across-breed and across-room scenarios, respectively. This study shows how feeding behavior and gut microbiota composition provide non-redundant information in predicting growth in swine.
Open AccessUse of Host Feeding Behavior and Gut Microbiome Data in Estimating Variance Components and Predicting Growth and Body Composition Traits in Swine
The purpose of this study was to investigate the use of feeding behavior in conjunction with gut microbiome sampled at two growth stages in predicting growth and body composition traits of finishing pigs. Six hundred and fifty-one purebred boars of three breeds: Duroc (DR), Landrace (LR), and Large White (LW), were studied. Feeding activities were recorded individually from 99 to 163 days of age. The 16S rRNA gene sequences were obtained from each pig at 123 ± 4 and 158 ± 4 days of age. When pigs reached market weight, body weight (BW), ultrasound backfat thickness (BF), ultrasound loin depth (LD), and ultrasound intramuscular fat (IMF) content were measured on live animals. Three models including feeding behavior (Model_FB), gut microbiota (Model_M), or both (Model_FB_M) as predictors, were investigated. Prediction accuracies were evaluated through cross-validation across genetic backgrounds using the leave-one-breed-out strategy and across rearing environments using the leave-one-room-out approach. The proportions of phenotypic variance of growth and body composition traits explained by feeding behavior ranged from 0.02 to 0.30, and from 0.20 to 0.52 when using gut microbiota composition. Overall prediction accuracy (averaged over traits and time points) of phenotypes was 0.24 and 0.33 for Model_FB, 0.27 and 0.19 for Model_M, and 0.40 and 0.35 for Model_FB_M for the across-breed and across-room scenarios, respectively. This study shows how feeding behavior and gut microbiota composition provide non-redundant information in predicting growth in swine.