Pasar al contenido principal

Impact of structural variation on the genome evolution of plant crops

RYC2022-037459-I

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Desarrollar, Atraer y Retener Talento
Subprograma Subprograma Estatal de Incorporación
Convocatoria Ramón y Cajal
Año convocatoria 2022
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario INSTITUT DE RECERCA I TECNOLOGIA AGROALIMENTARIES (IRTA)

Resultados relacionados

Resultados totales (Incluyendo duplicados): 12
Encontrada(s) 1 página(s)

Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Evaluation of deep learning for predicting rice traits using structural and single-nucleotide genomic variants

  • Vourlaki, Ioanna Theoni|||0000-0002-3381-5287
  • Ramos Onsins, Sebastián Ernesto|||0000-0002-1776-140X
  • Perez-Enciso, Miguel|||0000-0003-3524-995X
  • Castanera, Raúl|||0000-0002-3772-7727
Structural genomic variants (SVs) are prevalent in plant genomes and have played an important role in evolution and domestication, as they constitute a significant source of genomic and phenotypic variability. Nevertheless, most methods in quantitative genetics focusing on crop improvement, such as genomic prediction, consider only Single Nucleotide Polymorphisms (SNPs). Deep Learning (DL) is a promising strategy for genomic prediction, but its performance using SVs and SNPs as genetic markers remains unknown. We used rice to investigate whether combining SVs and SNPs can result in better trait prediction over SNPs alone and examine the potential advantage of Deep Learning (DL) networks over Bayesian Linear models. Specifically, the performances of BayesC (considering additive effects) and a Bayesian Reproducible Kernel Hilbert space (RKHS) regression (considering both additive and non-additive effects) were compared to those of two different DL architectures, the Multilayer Perceptron, and the Convolution Neural Network, to explore their prediction ability by using various marker input strategies. We found that exploiting structural and nucleotide variation slightly improved prediction ability on complex traits in 87% of the cases. DL models outperformed Bayesian models in 75% of the studied cases, considering the four traits and the two validation strategies used. Finally, DL systematically improved prediction ability of binary traits against the Bayesian models. Our study reveals that the use of structural genomic variants can improve trait prediction in rice, independently of the methodology used. Also, our results suggest that Deep Learning (DL) networks can perform better than Bayesian models in the prediction of binary traits, and in quantitative traits when the training and target sets are not closely related. This highlights the potential of DL to enhance crop improvement in specific scenarios and the importance to consider SVs in addition to SNPs in genomic selection.
Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Tandem LTR-retrotransposon structures are common and highly polymorphic in plant genomes

  • Morales-Díaz, Noemia
  • Sushko, Svitlana
  • Campos-Dominguez, Lucía|||0000-0002-8998-3394
  • Kopalli, Venkataramana
  • Golicz, Agnieszka A.
  • Castanera, Raúl|||0000-0002-3772-7727
  • Casacuberta, Josep M.|||0000-0002-5609-4152
LTR-retrotransposons (LTR-RT) are a major component of plant genomes and important drivers of genome evolution. Most LTR-RT copies in plant genomes are defective elements found as truncated copies, nested insertions or as part of more complex structures. The recent availability of highly contiguous plant genome assemblies based on long-read sequences now allows to perform detailed characterization of these complex structures and to evaluate their importance for plant genome evolution. The detailed analysis of two rice loci containing complex LTR-RT structures showed that they consist of tandem arrays of LTR copies sharing internal LTRs. Our analyses suggests that these LTR-RT tandems are the result of a single insertion and not of the recombination of two independent LTR-RT elements. Our results also suggest that gypsy elements may be more prone to form these structures. We show that these structures are highly polymorphic in rice and therefore have the potential to generate genetic variability. We have developed a computational pipeline (IDENTAM) that scans genome sequences and identifies tandem LTR-RT candidates. Using this tool, we have detected 266 tandems in a pangenome built from the genomes of 76 accessions of cultivated and wild rice, showing that tandem LTR-RT structures are frequent and highly polymorphic in rice. Running IDENTAM in the Arabidopsis, almond and cotton genomes showed that LTR-RT tandems are frequent in plant genomes of different size, complexity and ploidy level. The complexity of differentiating intra-element variations at the nucleotide level among haplotypes is very high, and we found that graph-based pangenomic methodologies are appropriate to resolve these structures. Our results show that LTR-RT elements can form tandem arrays. These structures are relatively abundant and highly polymorphic in rice and are widespread in the plant kingdom. Future studies will contribute to understanding how these structures originate and whether the variability that they generate has a functional impact.
Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Differential LTR‑retrotransposon dynamics across polyploidization, speciation, domestication, and improvement of cotton (Gossypium)

  • Campos-Dominguez, Lucía|||0000-0002-8998-3394
  • Castanera, Raúl|||0000-0002-3772-7727
  • Grover, Corrinne E.
  • Wendel, Jonathan|||0000-0003-2258-5081
  • Casacuberta, Josep M.|||0000-0002-5609-4152
Background; transposable elements are major components of plant genomes and major drivers of plant genome evolution. The cotton genus (Gossypium) is an excellent evolutionary model for polyploidization, speciation, domestication, and crop improvement. Here, we implement genome and pangenome analyses to study in detail the dynamics of LTR-retrotransposons during the cotton evolution. - Results we show that some LTR-retrotransposon lineages amplified in tetraploid cotton compared to their diploid progenitors, whereas others stayed stable or amplified but were removed through solo-LTR formation. Using species-level pangenomes we show that only a few lineages (CRM, Tekay, Ivana, and Tork) remained active after polyploidization and are still transposing. Tekay and CRM elements have re-shaped the centromeric and pericentromeric regions of tetraploid cottons in a subgenome specific manner, through new insertions but also selective eliminations through solo-LTR formation. On the other hand, Ivana and Tork have actively inserted within or close to genes affecting their expression. Finally, population-level analyses using the two pangenomes and data from 283 and 223 varieties of G. hirsutum and G. barbadense reveal changes in Transposon Insertion Polymorphism frequencies accompanying domestication and improvement of both species, suggesting the possibility of selection on linked regions. - Conclusions: our findings reveal that LTR-retrotransposon lineages followed differential dynamics during cotton evolution, displaying differences among species and the two coresident genomes of allopolyploid cotton. A handful of the LTR-retrotransposon lineages that expanded after polyploidization helped shape the genomes of both G. hirsutum and G. barbadense, impacting their centromere and pericentromeric regions as well as protein-coding genes.
IRTA Pubpro. Open Digital Archive
Publicación Artículo científico (JournalArticle)

Tandem LTR-retrotransposon structures are common and highly polymorphic in plant genomes

  • Morales-Díaz, Noemia
  • Sushko, Svitlana
  • Campos-Dominguez, Lucía
  • Kopalli, Venkataramana
  • Golicz, Agnieszka A.
  • Castanera, Raúl
  • Casacuberta, Josep M.
Background LTR-retrotransposons (LTR-RT) are a major component of plant genomes and important drivers of genome evolution. Most LTR-RT copies in plant genomes are defective elements found as truncated copies, nested insertions or as part of more complex structures. The recent availability of highly contiguous plant genome assemblies based on long-read sequences now allows to perform detailed characterization of these complex structures and to evaluate their importance for plant genome evolution. Results The detailed analysis of two rice loci containing complex LTR-RT structures showed that they consist of tandem arrays of LTR copies sharing internal LTRs. Our analyses suggests that these LTR-RT tandems are the result of a single insertion and not of the recombination of two independent LTR-RT elements. Our results also suggest that gypsy elements may be more prone to form these structures. We show that these structures are highly polymorphic in rice and therefore have the potential to generate genetic variability. We have developed a computational pipeline (IDENTAM) that scans genome sequences and identifies tandem LTR-RT candidates. Using this tool, we have detected 266 tandems in a pangenome built from the genomes of 76 accessions of cultivated and wild rice, showing that tandem LTR-RT structures are frequent and highly polymorphic in rice. Running IDENTAM in the Arabidopsis, almond and cotton genomes showed that LTR-RT tandems are frequent in plant genomes of different size, complexity and ploidy level. The complexity of differentiating intra-element variations at the nucleotide level among haplotypes is very high, and we found that graph-based pangenomic methodologies are appropriate to resolve these structures. Conclusions Our results show that LTR-RT elements can form tandem arrays. These structures are relatively abundant and highly polymorphic in rice and are widespread in the plant kingdom. Future studies will contribute to understanding how these structures originate and whether the variability that they generate has a functional impact., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. The work done at CRAG was funded by grant PID2022-143167NB-I00 funded by MICIU/AEI/ 10.13039/50110001103 and by “ERDF/EU” and grant CEX2019-000902-S funded by MICIU/AEI /10.13039/501100011033. NMD is funded by Grant PRE2020-095111 Funded by MCIU/AEI /10.13039/501100011033 and by “ESF Investing in your future”, and RC was partially funded by a Juan de la Cierva contract, grant IJC2020-045949-I funded by MICIU/AEI /10.13039/501100011033 and by European Union NextGenerationEU/PRTR”, and is now a Ramón y Cajal contract holder, RYC2022-037459-I funded by MICIU/AEI/ 10.13039/501100011033 and by FSE+. AAG was supported by the LOEWE Start Professorship from the Hessian Ministry for Science and the Arts. VK was supported by GRK 2843 from the German Research Foundation (DFG)., info:eu-repo/semantics/publishedVersion
IRTA Pubpro. Open Digital Archive
Publicación Artículo científico (JournalArticle)

Cultivar-Dependent Variations in the Microbiome of Grapevine Leaves

  • Castanera, Raúl
  • González-Miguel, Víctor M.
  • Escolà, Glòria
  • Olivé, Marta
  • Teixidó, Neus
  • Savé, Robert
  • Casacuberta, Josep María
  • San Segundo, Blanca
The grapevine (Vitis vinifera) is a major fruit crop of economic importance worldwide. Commercial grapevine cultivars are susceptible to infection by pathogenic microorganisms that cause diseases both in leaves and fruits, and it is known that the leaf microbiome plays an important role in plant health and fitness. In this study, shotgun metagenomic sequencing was used to characterize the microbial communities associated with grapevine leaves in three commercial varieties, Cabernet Sauvignon, Garnacha, and Marselan, grown in the same biogeographical unit. Metagenomic data revealed a differential enrichment of the microbial communities living inside grapevine leaves or on the leaf surface in the three varieties. The most abundant fungal taxa associated with grapevine leaves belong to the phylum Ascomycota, which included relevant pathogenic fungi for grapevines, such as Botrytis cinerea, Sclerotinia sclerotium, and Alternaria alternata, as well as several fungal species potentially pathogenic for grapevines (e.g., members of the Colletotrichum, Aspergillus, and Penicillium genera). Basidiomycota constituted a minor fraction of the fungal microbial communities. Grapevine leaves also harbored a diversity of bacterial taxa. At the phylum level, bacterial communities in all three varieties were primarily composed of Pseudomonadata, Bacillota, Bacteroidota, and a lower proportion of Actinomycetota. Differences in the fungal and bacterial community structures were observed between varieties, although they were more important in fungi. In particular, S. sclerotiorum and B. cinerea were found to preferentially colonize leaves in the Marselan and Garnacha varieties, respectively. These findings further support that the host genotype can shape its own microbiome in grapevines. A better understanding of the leaf microbiome in grapevines will provide the basis for the development of tailored strategies to prevent diseases in vineyards while helping to increase sustainability in grapevine production., This research was supported by grants PID2021-128825OB-I00 funded by MICIU/AEI/10.13039/501100011033 and by “ERDF/EU” to B.SS. and PID2019-106374RB-I00 funded by MICIU/AEI/10.13039/501100011033 and PID2022-143167NB-I00 funded by MICIU/AEI/10.13039/501100011033 and by “ERDF/EU grants to JC. We acknowledge financial support from CEX2019-000902-S funded by MICIU/AEI/10.13039/501100011033 through the “Severo Ochoa Program for Centres of Excellence in R&D” and the CERCA Program “Generalitat de Catalunya”. We acknowledge support of the publication fee by the CSIC Open Access Publication Support Initiative through its Unit of Information Resources for Research (URICI). We also thank the financial support from the Generalitat de Catalunya (grants 2021SGR00875 and 2021SGR01477). R.C was a recipient of a Juan de la Cierva grant from the Spanish Ministry of Science and Innovation (IJC2020-045949-I) funded by MICIU/AEI/10.13039/501100011033 and by the “European Union Next GenerationEU/PRTR” and a Ramon y Cajal fellowship (RYC2022-037459-I) funded by MICIU/AEI/10.13039/501100011033 and by ESF+., info:eu-repo/semantics/publishedVersion
IRTA Pubpro. Open Digital Archive
Publicación Artículo científico (JournalArticle)

Differential LTR‑retrotransposon dynamics across polyploidization, speciation, domestication, and improvement of cotton (Gossypium)

  • Campos-Dominguez, Lucía
  • Castanera, Raúl
  • Grover, Corrine E.
  • Wendel, Jonathan F.
  • Casacuberta, Josep M.
Background Transposable elements are major components of plant genomes and major drivers of plant genome evolution. The cotton genus (Gossypium) is an excellent evolutionary model for polyploidization, speciation, domestication, and crop improvement. Here, we implement genome and pangenome analyses to study in detail the dynamics of LTR-retrotransposons during the cotton evolution. Results We show that some LTR-retrotransposon lineages amplified in tetraploid cotton compared to their diploid progenitors, whereas others stayed stable or amplified but were removed through solo-LTR formation. Using species-level pangenomes we show that only a few lineages (CRM, Tekay, Ivana, and Tork) remained active after polyploidization and are still transposing. Tekay and CRM elements have re-shaped the centromeric and pericentromeric regions of tetraploid cottons in a subgenome specific manner, through new insertions but also selective eliminations through solo-LTR formation. On the other hand, Ivana and Tork have actively inserted within or close to genes affecting their expression. Finally, population-level analyses using the two pangenomes and data from 283 and 223 varieties of G. hirsutum and G. barbadense reveal changes in Transposon Insertion Polymorphism frequencies accompanying domestication and improvement of both species, suggesting the possibility of selection on linked regions. Conclusions Our findings reveal that LTR-retrotransposon lineages followed differential dynamics during cotton evolution, displaying differences among species and the two coresident genomes of allopolyploid cotton. A handful of the LTR-retrotransposon lineages that expanded after polyploidization helped shape the genomes of both G. hirsutum and G. barbadense, impacting their centromere and pericentromeric regions as well as protein-coding genes., This work was supported by PID2022-143167NB-I00 grant funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU to JC. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 945043 through the Postdoctoral Fellowship awarded to L.C.-D. by the AGenT H2020-MSCA-COFUND-2019 programme co-funded by the grant CEX2019-000902-S funded by MICIU/AEI/10.13039/501100011033. RC’s work was supported by his Ramón y Cajal fellowship (RYC2022-037459-I) funded by MICIU/AEI/10.13039/501100011033 and by ESF +. We acknowledge support on cotton research and polyploidy from the National Science Foundation, and we would also like to acknowledge Iowa State University Research IT for computational support., info:eu-repo/semantics/publishedVersion
IRTA Pubpro. Open Digital Archive
Publicación Artículo científico (JournalArticle)

Evaluation of deep learning for predicting rice traits using structural and single-nucleotide genomic variants

  • Vourlaki, Ioanna-Theoni
  • Ramos-Onsins, Sebastián E.
  • Pérez-Enciso, Miguel
  • Castanera, Raúl
Background - Structural genomic variants (SVs) are prevalent in plant genomes and have played an important role in evolution and domestication, as they constitute a signicant source of genomic and phenotypic variability. Never‑theless, most methods in quantitative genetics focusing on crop improvement, such as genomic prediction, consider only Single Nucleotide Polymorphisms (SNPs). Deep Learning (DL) is a promising strategy for genomic prediction, but its performance using SVs and SNPs as genetic markers remains unknown. Results - We used rice to investigate whether combining SVs and SNPs can result in better trait prediction over SNPs alone and examine the potential advantage of Deep Learning (DL) networks over Bayesian Linear models. Specically, the performances of BayesC (considering additive eects) and a Bayesian Reproducible Kernel Hilbert space (RKHS) regression (considering both additive and non‑additive eects) were compared to those of two dierent DL architectures, the Multilayer Perceptron, and the Convolution Neural Network, to explore their prediction ability by using vari‑ ous marker input strategies. We found that exploiting structural and nucleotide variation slightly improved prediction ability on complex traits in 87% of the cases. DL models outperformed Bayesian models in 75% of the studied cases, considering the four traits and the two validation strategies used. Finally, DL systematically improved prediction ability of binary traits against the Bayesian models. Conclusions - Our study reveals that the use of structural genomic variants can improve trait prediction in rice, independently of the methodology used. Also, our results suggest that Deep Learning (DL) networks can perform better than Bayesian models in the prediction of binary traits, and in quantitative traits when the training and target sets are not closely related. This highlights the potential of DL to enhance crop improvement in specific scenarios and the importance to consider SVs in addition to SNPs in genomic selection., Grants PID2019-108829RB-I00 and CEX2019-000902-S funded by MICIU/AEI/https://doi.org/10.13039/501100011033. IJC2020-045949-I and BES-2017-081139 funded by MICIU/AEI/ https://doi.org/10.13039/501100011033 and by “ESF Investing in your future”. RYC2022-037459-I funded by MICIU/AEI/ https://doi.org/10.13039/501100011033 and by “FSE+ , info:eu-repo/semantics/publishedVersion
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Evaluation of deep learning for predicting rice traits using structural and single-nucleotide genomic variants

  • Vourlaki, Ioanna-Theoni
  • Ramos-Onsins, Sebastian E.
  • Pérez-Enciso, Miguel
  • Castanera, Raúl
Structural genomic variants (SVs) are prevalent in plant genomes and have played an important role in evolution and domestication, as they constitute a significant source of genomic and phenotypic variability. Nevertheless, most methods in quantitative genetics focusing on crop improvement, such as genomic prediction, consider only Single Nucleotide Polymorphisms (SNPs). Deep Learning (DL) is a promising strategy for genomic prediction, but its performance using SVs and SNPs as genetic markers remains unknown., Grants PID2019-108829RB-I00 and CEX2019-000902-S funded by MICIU/AEI/https://doi.org/10.13039/501100011033. IJC2020-045949-I and BES-2017-081139 funded by MICIU/AEI/ https://doi.org/10.13039/501100011033 and by “ESF Investing in your future”. RYC2022-037459-I funded by MICIU/AEI/ https://doi.org/10.13039/501100011033 and by “FSE+ , With funding from the Spanish goverment through the "Severo Ochoa Centre of Excellence" accreditation (CEX2019-000902-S), Peer reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Tandem LTR-retrotransposon structures are common and highly polymorphic in plant genomes

  • Morales-Díaz, Noemia
  • Sushko, Svitlana
  • Campos-Dominguez, Lucía
  • Kopalli, Venkataramana
  • Golicz, Agnieszka A.
  • Castanera, Raúl
  • Casacuberta, Josep M.
Background LTR-retrotransposons (LTR-RT) are a major component of plant genomes and important drivers of genome evolution. Most LTR-RT copies in plant genomes are defective elements found as truncated copies, nested insertions or as part of more complex structures. The recent availability of highly contiguous plant genome assemblies based on long-read sequences now allows to perform detailed characterization of these complex structures and to evaluate their importance for plant genome evolution. Results The detailed analysis of two rice loci containing complex LTR-RT structures showed that they consist of tandem arrays of LTR copies sharing internal LTRs. Our analyses suggests that these LTR-RT tandems are the result of a single insertion and not of the recombination of two independent LTR-RT elements. Our results also suggest that gypsy elements may be more prone to form these structures. We show that these structures are highly polymorphic in rice and therefore have the potential to generate genetic variability. We have developed a computational pipeline (IDENTAM) that scans genome sequences and identifies tandem LTR-RT candidates. Using this tool, we have detected 266 tandems in a pangenome built from the genomes of 76 accessions of cultivated and wild rice, showing that tandem LTR-RT structures are frequent and highly polymorphic in rice. Running IDENTAM in the Arabidopsis, almond and cotton genomes showed that LTR-RT tandems are frequent in plant genomes of different size, complexity and ploidy level. The complexity of differentiating intra-element variations at the nucleotide level among haplotypes is very high, and we found that graph-based pangenomic methodologies are appropriate to resolve these structures. Conclusions Our results show that LTR-RT elements can form tandem arrays. These structures are relatively abundant and highly polymorphic in rice and are widespread in the plant kingdom. Future studies will contribute to understanding how these structures originate and whether the variability that they generate has a functional impact., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. The work done at CRAG was funded by grant PID2022-143167NB-I00 funded by MICIU/AEI/ 10.13039/50110001103 and by “ERDF/EU” and grant CEX2019-000902-S funded by MICIU/AEI /10.13039/501100011033. NMD is funded by Grant PRE2020-095111 Funded by MCIU/AEI /10.13039/501100011033 and by “ESF Investing in your future”, and RC was partially funded by a Juan de la Cierva contract, grant IJC2020-045949-I funded by MICIU/AEI /10.13039/501100011033 and by European Union NextGenerationEU/PRTR”, and is now a Ramón y Cajal contract holder, RYC2022-037459-I funded by MICIU/AEI/ 10.13039/501100011033 and by FSE+. AAG was supported by the LOEWE Start Professorship from the Hessian Ministry for Science and the Arts. VK was supported by GRK 2843 from the German Research Foundation (DFG)., info:eu-repo/semantics/publishedVersion
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Differential LTR‑retrotransposon dynamics across polyploidization, speciation, domestication, and improvement of cotton (Gossypium)

  • Campos-Dominguez, Lucía
  • Castanera, Raúl
  • Grover, Corrine E.
  • Wendel, Jonathan F.
  • Casacuberta, Josep M.
Background Transposable elements are major components of plant genomes and major drivers of plant genome evolution. The cotton genus (Gossypium) is an excellent evolutionary model for polyploidization, speciation, domestication, and crop improvement. Here, we implement genome and pangenome analyses to study in detail the dynamics of LTR-retrotransposons during the cotton evolution. Results We show that some LTR-retrotransposon lineages amplified in tetraploid cotton compared to their diploid progenitors, whereas others stayed stable or amplified but were removed through solo-LTR formation. Using species-level pangenomes we show that only a few lineages (CRM, Tekay, Ivana, and Tork) remained active after polyploidization and are still transposing. Tekay and CRM elements have re-shaped the centromeric and pericentromeric regions of tetraploid cottons in a subgenome specific manner, through new insertions but also selective eliminations through solo-LTR formation. On the other hand, Ivana and Tork have actively inserted within or close to genes affecting their expression. Finally, population-level analyses using the two pangenomes and data from 283 and 223 varieties of G. hirsutum and G. barbadense reveal changes in Transposon Insertion Polymorphism frequencies accompanying domestication and improvement of both species, suggesting the possibility of selection on linked regions. Conclusions Our findings reveal that LTR-retrotransposon lineages followed differential dynamics during cotton evolution, displaying differences among species and the two coresident genomes of allopolyploid cotton. A handful of the LTR-retrotransposon lineages that expanded after polyploidization helped shape the genomes of both G. hirsutum and G. barbadense, impacting their centromere and pericentromeric regions as well as protein-coding genes., This work was supported by PID2022-143167NB-I00 grant funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU to JC. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 945043 through the Postdoctoral Fellowship awarded to L.C.-D. by the AGenT H2020-MSCA-COFUND-2019 programme co-funded by the grant CEX2019-000902-S funded by MICIU/AEI/10.13039/501100011033. RC’s work was supported by his Ramón y Cajal fellowship (RYC2022-037459-I) funded by MICIU/AEI/10.13039/501100011033 and by ESF +. We acknowledge support on cotton research and polyploidy from the National Science Foundation, and we would also like to acknowledge Iowa State University Research IT for computational support., info:eu-repo/semantics/publishedVersion
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Evaluation of deep learning for predicting rice traits using structural and single-nucleotide genomic variants

  • Vourlaki, Ioanna-Theoni
  • Ramos-Onsins, Sebastián E.
  • Pérez-Enciso, Miguel
  • Castanera, Raúl
Background - Structural genomic variants (SVs) are prevalent in plant genomes and have played an important role in evolution and domestication, as they constitute a signicant source of genomic and phenotypic variability. Never‑theless, most methods in quantitative genetics focusing on crop improvement, such as genomic prediction, consider only Single Nucleotide Polymorphisms (SNPs). Deep Learning (DL) is a promising strategy for genomic prediction, but its performance using SVs and SNPs as genetic markers remains unknown. Results - We used rice to investigate whether combining SVs and SNPs can result in better trait prediction over SNPs alone and examine the potential advantage of Deep Learning (DL) networks over Bayesian Linear models. Specically, the performances of BayesC (considering additive eects) and a Bayesian Reproducible Kernel Hilbert space (RKHS) regression (considering both additive and non‑additive eects) were compared to those of two dierent DL architectures, the Multilayer Perceptron, and the Convolution Neural Network, to explore their prediction ability by using vari‑ ous marker input strategies. We found that exploiting structural and nucleotide variation slightly improved prediction ability on complex traits in 87% of the cases. DL models outperformed Bayesian models in 75% of the studied cases, considering the four traits and the two validation strategies used. Finally, DL systematically improved prediction ability of binary traits against the Bayesian models. Conclusions - Our study reveals that the use of structural genomic variants can improve trait prediction in rice, independently of the methodology used. Also, our results suggest that Deep Learning (DL) networks can perform better than Bayesian models in the prediction of binary traits, and in quantitative traits when the training and target sets are not closely related. This highlights the potential of DL to enhance crop improvement in specific scenarios and the importance to consider SVs in addition to SNPs in genomic selection., Grants PID2019-108829RB-I00 and CEX2019-000902-S funded by MICIU/AEI/https://doi.org/10.13039/501100011033. IJC2020-045949-I and BES-2017-081139 funded by MICIU/AEI/ https://doi.org/10.13039/501100011033 and by “ESF Investing in your future”. RYC2022-037459-I funded by MICIU/AEI/ https://doi.org/10.13039/501100011033 and by “FSE+ , info:eu-repo/semantics/publishedVersion
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Cultivar-Dependent Variations in the Microbiome of Grapevine Leaves

  • Castanera, Raúl
  • González-Miguel, Víctor M.
  • Escolà, Glòria
  • Olivé, Marta
  • Teixidó, Neus
  • Savé, Robert
  • Casacuberta, Josep María
  • San Segundo, Blanca
The grapevine (Vitis vinifera) is a major fruit crop of economic importance worldwide. Commercial grapevine cultivars are susceptible to infection by pathogenic microorganisms that cause diseases both in leaves and fruits, and it is known that the leaf microbiome plays an important role in plant health and fitness. In this study, shotgun metagenomic sequencing was used to characterize the microbial communities associated with grapevine leaves in three commercial varieties, Cabernet Sauvignon, Garnacha, and Marselan, grown in the same biogeographical unit. Metagenomic data revealed a differential enrichment of the microbial communities living inside grapevine leaves or on the leaf surface in the three varieties. The most abundant fungal taxa associated with grapevine leaves belong to the phylum Ascomycota, which included relevant pathogenic fungi for grapevines, such as Botrytis cinerea, Sclerotinia sclerotium, and Alternaria alternata, as well as several fungal species potentially pathogenic for grapevines (e.g., members of the Colletotrichum, Aspergillus, and Penicillium genera). Basidiomycota constituted a minor fraction of the fungal microbial communities. Grapevine leaves also harbored a diversity of bacterial taxa. At the phylum level, bacterial communities in all three varieties were primarily composed of Pseudomonadata, Bacillota, Bacteroidota, and a lower proportion of Actinomycetota. Differences in the fungal and bacterial community structures were observed between varieties, although they were more important in fungi. In particular, S. sclerotiorum and B. cinerea were found to preferentially colonize leaves in the Marselan and Garnacha varieties, respectively. These findings further support that the host genotype can shape its own microbiome in grapevines. A better understanding of the leaf microbiome in grapevines will provide the basis for the development of tailored strategies to prevent diseases in vineyards while helping to increase sustainability in grapevine production., This research was supported by grants PID2021-128825OB-I00 funded by MICIU/AEI/10.13039/501100011033 and by “ERDF/EU” to B.SS. and PID2019-106374RB-I00 funded by MICIU/AEI/10.13039/501100011033 and PID2022-143167NB-I00 funded by MICIU/AEI/10.13039/501100011033 and by “ERDF/EU grants to JC. We acknowledge financial support from CEX2019-000902-S funded by MICIU/AEI/10.13039/501100011033 through the “Severo Ochoa Program for Centres of Excellence in R&D” and the CERCA Program “Generalitat de Catalunya”. We acknowledge support of the publication fee by the CSIC Open Access Publication Support Initiative through its Unit of Information Resources for Research (URICI). We also thank the financial support from the Generalitat de Catalunya (grants 2021SGR00875 and 2021SGR01477). R.C was a recipient of a Juan de la Cierva grant from the Spanish Ministry of Science and Innovation (IJC2020-045949-I) funded by MICIU/AEI/10.13039/501100011033 and by the “European Union Next GenerationEU/PRTR” and a Ramon y Cajal fellowship (RYC2022-037459-I) funded by MICIU/AEI/10.13039/501100011033 and by ESF+., info:eu-repo/semantics/publishedVersion