A directed learning strategy integrating multiple omic data improves genomic prediction

Hu, Xuehai; Xie, Weibo; Wu, Chengchao; Xu, Shizhong

doi:10.1111/pbi.13117

Cited by 36 publications

(28 citation statements)

References 43 publications

(68 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This makes it possible to uncover genotype-phenotype relationships using different types of data. Related studies were reported using omics data to perform genomic prediction for complex traits in humans [19,20], plants [21][22][23][24], and model animals [25,26]. Most of these studies focused on integrating multiple omics data into a prediction model to improve prediction accuracy [22,[25][26][27].…”

Section: Introductionmentioning

confidence: 99%

Multi-omics-data-assisted genomic feature markers preselection improves the accuracy of genomic prediction

Ye¹,

Li²,

Zhang³

2020

J Animal Sci Biotechnol

View full text Add to dashboard Cite

Background Presently, multi-omics data (e.g., genomics, transcriptomics, proteomics, and metabolomics) are available to improve genomic predictors. Omics data not only offers new data layers for genomic prediction but also provides a bridge between organismal phenotypes and genome variation that cannot be readily captured at the genome sequence level. Therefore, using multi-omics data to select feature markers is a feasible strategy to improve the accuracy of genomic prediction. In this study, simultaneously using whole-genome sequencing (WGS) and gene expression level data, four strategies for single-nucleotide polymorphism (SNP) preselection were investigated for genomic predictions in the Drosophila Genetic Reference Panel. Results Using genomic best linear unbiased prediction (GBLUP) with complete WGS data, the prediction accuracies were 0.208 ± 0.020 (0.181 ± 0.022) for the startle response and 0.272 ± 0.017 (0.307 ± 0.015) for starvation resistance in the female (male) lines. Compared with GBLUP using complete WGS data, both GBLUP and the genomic feature BLUP (GFBLUP) did not improve the prediction accuracy using SNPs preselected from complete WGS data based on the results of genome-wide association studies (GWASs) or transcriptome-wide association studies (TWASs). Furthermore, by using SNPs preselected from the WGS data based on the results of the expression quantitative trait locus (eQTL) mapping of all genes, only the startle response had greater accuracy than GBLUP with the complete WGS data. The best accuracy values in the female and male lines were 0.243 ± 0.020 and 0.220 ± 0.022, respectively. Importantly, by using SNPs preselected based on the results of the eQTL mapping of significant genes from TWAS, both GBLUP and GFBLUP resulted in great accuracy and small bias of genomic prediction. Compared with the GBLUP using complete WGS data, the best accuracy values represented increases of 60.66% and 39.09% for the starvation resistance and 27.40% and 35.36% for startle response in the female and male lines, respectively. Conclusions Overall, multi-omics data can assist genomic feature preselection and improve the performance of genomic prediction. The new knowledge gained from this study will enrich the use of multi-omics in genomic prediction.

show abstract

Section: Introductionmentioning

confidence: 99%

Multi-omics-data-assisted genomic feature markers preselection improves the accuracy of genomic prediction

Ye¹,

Li²,

Zhang³

2020

J Animal Sci Biotechnol

View full text Add to dashboard Cite

show abstract

“…Both this study and several previous studies have indicated that integrating transcriptomic data into genomic prediction is a feasible method to improve the power of genomic prediction (Azodi et al, 2019;Hu et al, 2019;Morgante et al, 2019).…”

Section: Challenges For Integrating Transcriptomic Data Into Genomic mentioning

confidence: 71%

“…Related studies were reported using omics data to perform genomic prediction for complex traits in humans (Vazquez et al, 2016;Dimitrakopoulos et al, 2017), plants (Xu et al, 2017;Azodi et al, 2019;Hu et al, 2019;Wang et al, 2019), and model animals (Li et al, 2019;Morgante et al, 2019). Most of these studies focused on integrating multiple omics data into a prediction model to improve prediction accuracy (Guo et al, 2016;Xu et al, 2017;Li et al, 2019;Morgante et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Multi-omics-data-assisted genomic feature preselection improves the accuracy of genomic prediction

Zhang

2020

Preprint

View full text Add to dashboard Cite

Background: Presently, multi-omics data (e.g., genomics, transcriptomics, proteomics, and metabolomics) are available for genomic prediction. Omics data not only offer new data layers for genomic prediction but also provide a bridge between organismal phenotypes and genome variation that cannot be readily captured at the genome sequence level. Therefore, using multi-omics data to select feature markers is a feasible strategy to improve the accuracy of genomic prediction. In this study, simultaneously using whole-genome sequencing (WGS) and gene expression level data, four strategies for single-nucleotide polymorphism (SNP) preselection were investigated for genomic predictions in the Drosophila Genetic Reference Panel.Results: Using genomic best linear unbiased prediction (GBLUP) with complete WGS data, the prediction accuracy values were 0.208±0.020 (0.181±0.022) for the startle response and 0.272±0.017 (0.307±0.015) for starvation resistance in the female (male) lines. Compared with GBLUP using complete WGS data, both GBLUP and the genomic feature BLUP (GFBLUP) did not improve the prediction accuracy using SNPs preselected from the complete WGS data based on the results of genome-wide association studies (GWASs) or transcriptome-wide association studies (TWASs). Furthermore, by using SNPs preselected from the WGS data based on the results of the expression quantitative trait locus (eQTL) mapping of all genes, only the startle response had greater accuracy than GBLUP with the complete WGS data. The best accuracy values in the female and male lines were 0.243±0.020 and 0.220±0.022, respectively. Importantly, by using SNPs preselected based on the results of the eQTL mapping of significant genes from TWAS, both GBLUP and GFBLUP resulted in a greater accuracy and smaller bias of genomic prediction. For the startle response, the best accuracy values were 0.258±0.019 (0.237±0.019) for GBLUP and 0.265±0.018 (0.245±0.020) for GFBLUP in the female (male) lines. For starvation resistance, the best accuracy values were 0.437±0.015 (0.427±0.015) for GBLUP and 0.419±0.016 (0.390±0.014) for GFBLUP in female (male) lines. Compared to the GBLUP with complete WGS data, the best accuracy values represented increases of 60.66% and 39.09% for the startle response and 27.40% and 35.36% for starvation resistance in the female and male lines, respectively. Conclusions: Overall, multi-omics data can assist genomic feature preselection and improve the performance of genomic prediction. The new knowledge gained from this study will enrich the use of multi-omics in genomic prediction.

show abstract

“…Recently, Omics data started to be integrated into the genomic selection models. For example, transcriptomic and metabolomic data have been combined into genomic selection to boost the power of prediction (Hu et al 2019). Additionally, by incorporating evolutionary information into the genomic selection model, the prediction accuracy has been improved for up to 4% for yield-related traits in maize (Yang et al 2017).…”

Section: )mentioning

confidence: 99%

Genomic selection to optimize doubled haploid-based hybrid breeding in maize

Cheng

Guo

et al. 2020

Preprint

View full text Add to dashboard Cite

Crop improvement, as a long-term endeavor, requires continuous innovations in technique from multiple perspectives. Doubled haploid (DH) technology for pure inbred production, which shaves years off of the conventional selfing approach, has been widely used for breeding. However, the final success rate of in vivo maternal DH production is determined by four factors: haploids induction, haploids identification, chromosome doubling, and successful selfing of the fertile haploid plants to produce DH seeds. Traits in each of these steps, if they can be accurately predicted using genomic selection methods, will help adjust the DH production protocol and simplify the logistics and save costs. Here, a hybrid population (N=158) was generated based on an incomplete half diallel design using 27 elite inbred lines. These hybrids were induced to create F1-derived haploid families. The hybrid materials, as well as the 27 inbreds, the inbred-derived haploids (N=200), and the F1-derived haploids (N=5,000) were planted in the field to collect four DH-production traits, three yield-related traits, and three developmental traits. Quantitative genetics analysis suggested that in both diploids and haploid families, most of the developmental traits showed high heritability, while the DH-production and developmental traits exhibited intermediate levels of heritability. By employing different genomic selection models, our results showed that the prediction accuracy ranged from 0.52 to 0.59 for the DH-production traits, 0.50 to 0.68 for the yield-related traits, and 0.44 to 0.87 for the developmental traits. Further analysis using index selection achieved the highest prediction accuracy when considering both DH production efficiency and the agronomic trait performance. Furthermore, the long-term responses through simulation confirmed that index selection would increase the genetic gain for targeted agronomic traits while maintaining the DH production efficiency. Therefore, our study provides an optimization strategy to integrate GS technology for DH-based hybrid breeding.

show abstract

A directed learning strategy integrating multiple omic data improves genomic prediction

Cited by 36 publications

References 43 publications

Multi-omics-data-assisted genomic feature markers preselection improves the accuracy of genomic prediction

Multi-omics-data-assisted genomic feature markers preselection improves the accuracy of genomic prediction

Multi-omics-data-assisted genomic feature preselection improves the accuracy of genomic prediction

Genomic selection to optimize doubled haploid-based hybrid breeding in maize

Contact Info

Product

Resources

About