Search

Among the Mediterranean horticultural landraces, garlic is one of the crops most threatened by genetic erosion. Due to its sexual sterility and to the incidence of seed-borne diseases, historical varieties have been widely replaced by commercial cultivars. In Catalonia, despite the historical relevance of the crop, solely the Belltall garlic landrace is cultivated for commercial purposes. To assess the genotypic and phenotypic diversity within the Belltall garlic, we evaluated sixteen local accessions and five recognized traditional and modern varieties as controls. Genetic analysis with SSR and InDel markers showed low genetic diversity within the Belltall population, grouping modern and traditional varieties separately. Farmers and consumers were involved in the definition of the landrace ideotype and classified the materials by means of projective mapping. Scant phenotypic diversity was found within the Belltall landrace, which is characterized by its color profile and the small size of bulb and cloves. The Belltall landrace grown outside its area of origin lost the distinctive quality signals that differentiate the landrace from the commercial cultivars (clove appearance), indicating that the high quality of the landrace is under genotype-by-environment effects (i.e. local adaptation). Moreover, the size of the Belltall sowing clove had a strong effect on the harvested bulb size. Our research represents a case study for the description of the variability within garlic landraces and an approach to quantify the phenomenon of local adaptation that currently drives their conservation.

Among the Mediterranean horticultural landraces, garlic is one of the crops most threatened by genetic erosion. Due to its sexual sterility and to the incidence of seed-borne diseases, historical varieties have been widely replaced by commercial cultivars. In Catalonia, despite the historical relevance of the crop, solely the Belltall garlic landrace is cultivated for commercial purposes. To assess the genotypic and phenotypic diversity within the Belltall garlic, we evaluated sixteen local accessions and five recognized traditional and modern varieties as controls. Genetic analysis with SSR and InDel markers showed low genetic diversity within the Belltall population, grouping modern and traditional varieties separately. Farmers and consumers were involved in the definition of the landrace ideotype and classified the materials by means of projective mapping. Scant phenotypic diversity was found within the Belltall landrace, which is characterized by its color profile and the small size of bulb and cloves. The Belltall landrace grown outside its area of origin lost the distinctive quality signals that differentiate the landrace from the commercial cultivars (clove appearance), indicating that the high quality of the landrace is under genotype-by-environment effects (i.e. local adaptation). Moreover, the size of the Belltall sowing clove had a strong effect on the harvested bulb size. Our research represents a case study for the description of the variability within garlic landraces and an approach to quantify the phenomenon of local adaptation that currently drives their conservation., This research was supported by the Ajuntament de Passanant i Belltall, the Diputació de Tarragona, the Catalan Government through the Pla de la Biodiversitat Cultivada (grant ARP152/21/000010), and the grant ACCIÓ -Eurecat (Project PRIV2020-PLANTOMICS), This research was supported by the Ajuntament de Passanant i Belltall, the Diputació de Tarragona, the Catalan Government through the Pla de la Biodiversitat Cultivada (grant ARP152/21/000010), and the grant ACCIÓ -Eurecat (Project PRIV2020-PLANTOMICS), Postprint (published version)

Aquesta és una correcció a l'article 10.34133/2021/9812910

Cultivated strawberry, Fragaria × ananassa, has a complex aroma due to the presence of more than 350 volatile organic compounds (VOCs). However, a mixture of only 19 compounds, called Key Volatile Compounds (KVC), can impart the main strawberry aroma. The octoploid nature of the cultivated strawberry species (2n = 8x = 56) adds complexity to the heritance of the accumulation of the volatiles responsible for aroma. An F1 population cross between two breeding parental lines, FC50 and FD54, was phenotyped for aroma by SPME GCMS during six harvests. A total of 58 compounds were identified: 33 esters, nine terpenes, seven aldehydes, four lactones, two furans, one acid, one alkane and one alcohol, of which 16 were KVCs. A total of 179 QTLs were found, and 85 of these were detected in at least three harvests, of which 50 QTLs were considered major (LOD > 4.0) and detected in five or six analyzed harvests. Several clusters of ester QTLs associated with fruity aroma were discovered, such as QTLs for esters that share hexanoate group that were mapped in LG4A (Hexanoate_4A), those that share acetate and octyl groups in LG6A (Acetate_6A and Octyl_6A) or those with the same methyl group in LG7B (Methyl_7B). Different terpene QTLs associated with floral aroma appear grouped in a cluster in LG3C (Terpene_3C). Some of these clusters of QTLs were validated in a second F2 population, a cross of “Camarosa” and “Dover,” that was also phenotyped for three years. Selected SNPs from floral and fruity aroma QTLs were tested in a third population, which will most likely be useful for marker-assisted breeding (MAB).

Cultivated strawberry, Fragaria × ananassa, has a complex aroma due to the presence of more than 350 volatile organic compounds (VOCs). However, a mixture of only 19 compounds, called Key Volatile Compounds (KVC), can impart the main strawberry aroma. The octoploid nature of the cultivated strawberry species (2n = 8x = 56) adds complexity to the heritance of the accumulation of the volatiles responsible for aroma. An F1 population cross between two breeding parental lines, FC50 and FD54, was phenotyped for aroma by SPME GCMS during six harvests. A total of 58 compounds were identified: 33 esters, nine terpenes, seven aldehydes, four lactones, two furans, one acid, one alkane and one alcohol, of which 16 were KVCs. A total of 179 QTLs were found, and 85 of these were detected in at least three harvests, of which 50 QTLs were considered major (LOD > 4.0) and detected in five or six analyzed harvests. Several clusters of ester QTLs associated with fruity aroma were discovered, such as QTLs for esters that share hexanoate group that were mapped in LG4A (Hexanoate_4A), those that share acetate and octyl groups in LG6A (Acetate_6A and Octyl_6A) or those with the same methyl group in LG7B (Methyl_7B). Different terpene QTLs associated with floral aroma appear grouped in a cluster in LG3C (Terpene_3C). Some of these clusters of QTLs were validated in a second F2 population, a cross of “Camarosa” and “Dover,” that was also phenotyped for three years. Selected SNPs from floral and fruity aroma QTLs were tested in a third population, which will most likely be useful for marker-assisted breeding (MAB).

In the article titled “Automatic Fruit Morphology Phenome and Genetic Analysis: An Application in the Octoploid Strawberry” [1], some funding information was omitted...

Cultivated strawberry, Fragaria × ananassa, has a complex aroma due to the presence of more than 350 volatile organic compounds (VOCs). However, a mixture of only 19 compounds, called Key Volatile Compounds (KVC), can impart the main strawberry aroma. The octoploid nature of the cultivated strawberry species (2n = 8x = 56) adds complexity to the heritance of the accumulation of the volatiles responsible for aroma. An F1 population cross between two breeding parental lines, FC50 and FD54, was phenotyped for aroma by SPME GCMS during six harvests. A total of 58 compounds were identified: 33 esters, nine terpenes, seven aldehydes, four lactones, two furans, one acid, one alkane and one alcohol, of which 16 were KVCs. A total of 179 QTLs were found, and 85 of these were detected in at least three harvests, of which 50 QTLs were considered major (LOD > 4.0) and detected in five or six analyzed harvests. Several clusters of ester QTLs associated with fruity aroma were discovered, such as QTLs for esters that share hexanoate group that were mapped in LG4A (Hexanoate_4A), those that share acetate and octyl groups in LG6A (Acetate_6A and Octyl_6A) or those with the same methyl group in LG7B (Methyl_7B). Different terpene QTLs associated with floral aroma appear grouped in a cluster in LG3C (Terpene_3C). Some of these clusters of QTLs were validated in a second F2 population, a cross of “Camarosa” and “Dover,” that was also phenotyped for three years. Selected SNPs from floral and fruity aroma QTLs were tested in a third population, which will most likely be useful for marker-assisted breeding (MAB).

Genomic Selection (GS) is the procedure whereby molecular information is used to predict complex phenotypes and it is standard in many animal and plant breeding schemes. However, only a small number of studies have been reported in horticultural crops, and in polyploid species in particular. In this paper, we have developed a versatile forward simulation tool, called polyploid Sequence Based Virtual Breeding (pSBVB), to evaluate GS strategies in polyploids; pSBVB is an efficient gene dropping software that can simulate any number of complex phenotypes, allowing a very flexible modeling of phenotypes suited to polyploids. As input, it takes genotype data from the founder population, which can vary from single nucleotide polymorphisms (SNP) chips up to sequence, a list of causal variants for every trait and their heritabilities, and the pedigree. Recombination rates between homeologous chromosomes can be specified, so that both allo- and autopolyploid species can be considered. The program outputs phenotype and genotype data for all individuals in the pedigree. Optionally, it can produce several genomic relationship matrices that consider exact or approximate genotype values. pSBVB can therefore be used to evaluate GS strategies in polyploid species (say varying SNP density, genetic architecture or population size, among other factors), or to optimize experimental designs for association studies. We illustrate pSBVB with SNP data from tetraploid potato and partial sequence data from octoploid strawberry, and we show that GS is a promising breeding strategy for polyploid species but that the actual advantage critically depends on the underlying genetic architecture. Source code, examples and a complete manual are freely available in GitHub https://github.com/lauzingaretti/pSBVB.

Aquest article té una correcció a 10.34133/2022/9873618, Automatizing phenotype measurement will decisively contribute to increase plant breeding efficiency. Among phenotypes, morphological traits are relevant in many fruit breeding programs, as appearance influences consumer preference. Often, these traits are manually or semiautomatically obtained. Yet, fruit morphology evaluation can be enhanced using fully automatized procedures and digital images provide a cost-effective opportunity for this purpose. Here, we present an automatized pipeline for comprehensive phenomic and genetic analysis of morphology traits extracted from internal and external strawberry (Fragaria x ananassa) images. The pipeline segments, classifies, and labels the images and extracts conformation features, including linear (area, perimeter, height, width, circularity, shape descriptor, ratio between height and width) and multivariate (Fourier elliptical components and Generalized Procrustes) statistics. Internal color patterns are obtained using an autoencoder to smooth out the image. In addition, we develop a variational autoencoder to automatically detect the most likely number of underlying shapes. Bayesian modeling is employed to estimate both additive and dominance effects for all traits. As expected, conformational traits are clearly heritable. Interestingly, dominance variance is higher than the additive component for most of the traits. Overall, we show that fruit shape and color can be quickly and automatically evaluated and are moderately heritable. Although we study strawberry images, the algorithm can be applied to other fruits, as shown in the GitHub repository.

[CA]: L'agricultura moderna depèn àmpliament de l'ús de sofisticades eines informàtiques per analitzar dades massives, tant genotípiques com fenotípiques. La selecció genòmica (SG), que consisteix en predir característiques complexes utilitzant marcadors genètics d'ampli espectre, ha estat aprofitada pels milloradors de plantes i animals, al llarg de les últimes dècades, per produir un considerable augment del guany genètic, reduint el nombre de mostres a testar al camp. Paral·lelament, la implantació de l'electrònica, els sensors, les càmeres digitals, els vehicles aeris no tripulats i l'espectrometria de masses, entre d'altres, han obert una finestra d'oportunitats en l'àmbit de la ""fenómica"", augmentant ràpidament la quantitat de dades disponibles. Totes les tecnologies 'òmiques' també proporcionen noves fonts d'informació, permetent no només la caracterització del mateix organisme, sinó també seva del seu metagenoma. Per tant, un dels majors reptes actuals és combinar totes aquestes dades heterogènies per transformar-les en informació valuosa que ajudi els milloradors a prendre decisions eficaces. El present treball aborda una varietat de problemes de predicció genòmica i fenómica, tots ells amb l'objectiu comú d'explorar avantatges i desavantatges de l'ús de ML en agricultura. Les dues primeres contribucions tracten de problemes de predicció genòmica, mentre que els dos capítols següents s'ocupen de la fenómica, i l'última investigació tracta de la integració de dades. Els resultats obtinguts aquí mostren com les tècniques d'aprenentatge automàtic (ML) poden potenciar l'agricultura moderna en múltiples vies. No obstant això, queda molta feina per fer i es requereixen desenvolupaments específics de ML per potenciar el guany genètic en els programes de millora., [ES]: La agricultura moderna depende ampliamente del uso de sofisticadas herramientas informáticas para analizar datos masivos, tanto genotípicos como fenotípicos. La selección genómica (SG), que consiste en predecir rasgos complejos utilizando marcadores genéticos de amplio espectro, ha sido aprovechada por los mejoradores de plantas y animales a lo largo de las últimas décadas, para producir un considerable aumento de la ganancia genética, reduciendo el número de muestras a testear en el campo. Paralelamente, la implantación de la electrónica, los sensores, las cámaras digitales, los vehículos aéreos no tripulados y la espectrometría de masas, entre otros, han abierto una ventana de oportunidades en el ámbito de la ""fenómica"", aumentando rápidamente la cantidad de datos disponibles. Todas las tecnologías 'ómicas' también proporcionan nuevas fuentes de información, permitiendo no sólo la caracterización del propio organismo, sino también de su metagenoma. Por lo tanto, uno de los mayores desafíos actuales es combinar todos estos datos heterogéneos para transformarlos en información valiosa que ayude a los mejoradores a tomar decisiones eficaces. El presente trabajo aborda una variedad de problemas de predicción genómica y fenómica, todos ellos con el objetivo común de explorar ventajas y desventajas del uso de ML en agricultura. Las dos primeras contribuciones tratan de problemas de predicción genómica, mientras que los dos capítulos siguientes se ocupan de la fenómica, y la última investigación trata de la integración de datos. Los resultados obtenidos aquí muestran cómo las técnicas de aprendizaje automático (ML) pueden potenciar la agricultura moderna en múltiples vías. Sin embargo, queda mucho trabajo por hacer y se requieren desarrollos específicos de ML para potenciar la ganancia genética en los programas de mejora., [EN]: Modern agriculture relies heavily on sophisticated computational tools that involve genomics and phenomics data at a large scale. As for genomics, over the past few decades, plant and animal breeders have taken advantage of genomic selection (GS), which is the breeding strategy that consists of predicting complex traits using genomic wide genetic markers. GS has two main advantages over traditional approaches: increasing genetic gain and reducing the amount of data to be tested in the field. In parallel, the implementation of electronic, sensors, digital cameras, unmanned aerial vehicles, mass spectrometry, among others, have opened a window of opportunities in the ‘phenomics’ area, rapidly increasing the amount of available data. Phenotyping does not end here, as ‘omics’ technologies also provide a new source of information, allowing not only the characterization of the organism itself but also of its metagenome. The current challenge is to transform and combine all these heterogeneous data into valuable information that helps the breeder to make better and more effective decisions. The present work deals with a variety of genomic prediction and phenomics problems, all with the shared objective of exploring the strengths and weaknesses of ML techniques in agriculture. The first two contributions deal with genomic prediction issues while the following two chapters are concerned with phenomics, followed by the last research on data integration. The results obtained here show how machine learning (ML) techniques can empower modern agriculture in multiple avenues. However, much work remains to be done and specific ML developments are required to enhance genetic gain in breeding programs.

Plant architecture is defined by fates and positions of meristematic tissues and has direct consequences on yield potential and environmental adaptation of the plant. In strawberries (Fragaria vesca L. and F. × ananassa Duch.), shoot apical meristems can remain vegetative or differentiate into a terminal inflorescence meristem. Strawberry axillary buds (AXBs) are located in leaf axils and can either remain dormant or follow one of the two possible developmental fates. AXBs can either develop into stolons needed for clonal reproduction or into branch crowns (BCs) that can bear their own terminal inflorescences under favorable conditions. Although AXB fate has direct consequences on yield potential and vegetative propagation of strawberries, the regulation of AXB fate has so far remained obscure. We subjected a number of woodland strawberry (F. vesca L.) natural accessions and transgenic genotypes to different environmental conditions and growth regulator treatments to demonstrate that strawberry AXB fate is regulated either by environmental or endogenous factors, depending on the AXB position on the plant. We confirm that the F. vesca GIBBERELLIN20-oxidase4 (FvGA20ox4) gene is indispensable for stolon development and under tight environmental regulation. Moreover, our data show that apical dominance inhibits the outgrowth of the youngest AXB as BCs, although the effect of apical dominance can be overrun by the activity of FvGA20ox4. Finally, we demonstrate that the FvGA20ox4 is photoperiodically regulated via FvSOC1 (F. vesca SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1) at 18°C, but at higher temperature of 22°C an unidentified FvSOC1-independent pathway promotes stolon development.

The cultivated strawberry (Fragaria x ananassa) is an octoploid species (2n = 8x = 56), appreciated widely for its fruit. There have been very few studies on fruit quality traits, which are known to be mostly polygenic and environmentally dependent. To identify higher genetic variability, two non-related populations were genotyped: an F1 population cross between ‘FC50’ and ‘FD54’ and an F2 population cross between ‘Camarosa’ and ‘Dover’, hybridizing both with IStraw35k and IStraw90k SNP arrays, respectively. The F1 genetic map was constructed with 14595 SNPs and the F2 map with 7977 SNPs. High collinearity was observed when comparing one genetic map with the other and on comparing both with the octoploid genome. To assess fruit variability, both populations were phenotyped for shape, firmness, taste and other fruit traits over the 2016−2019 period. With QTL analyses, 33 stable QTLs were mapped in the ‘FC50xFD54’ population, and three hotspot regions were found for shape traits in LG3A, LG4D and LG6D. In the ‘21AF’ population, only eight stable QTLs were detected. Despite that, two major and stable QTLs were mapped in the same interval of confidence for both populations. A shared fruit shape ratio QTL which explained around 25 % of trait variance was mapped in LG3A, and a shared firmness QTL explaining 26.9 % of trait variance in LG7C. For the first time, two QTLs were discovered in LG3A and LG4A for a phenotype neck without achenes. When analysing two different mapping populations, in addition to finding specific QTL regions for the studied traits, a narrowing down of the interval of confidence for the shared QTLs is achieved. As a result of this study, a new set of SNPs for fruit firmness and shape is now available for use in MAS in strawberry breeding programs.

Automatizing phenotype measurement will decisively contribute to increase plant breeding efficiency. Among phenotypes, morphological traits are relevant in many fruit breeding programs, as appearance influences consumer preference. Often, these traits are manually or semiautomatically obtained. Yet, fruit morphology evaluation can be enhanced using fully automatized procedures and digital images provide a cost-effective opportunity for this purpose. Here, we present an automatized pipeline for comprehensive phenomic and genetic analysis of morphology traits extracted from internal and external strawberry (Fragaria x ananassa) images. The pipeline segments, classifies, and labels the images and extracts conformation features, including linear (area, perimeter, height, width, circularity, shape descriptor, ratio between height and width) and multivariate (Fourier elliptical components and Generalized Procrustes) statistics. Internal color patterns are obtained using an autoencoder to smooth out the image. In addition, we develop a variational autoencoder to automatically detect the most likely number of underlying shapes. Bayesian modeling is employed to estimate both additive and dominance effects for all traits. As expected, conformational traits are clearly heritable. Interestingly, dominance variance is higher than the additive component for most of the traits. Overall, we show that fruit shape and color can be quickly and automatically evaluated and are moderately heritable. Although we study strawberry images, the algorithm can be applied to other fruits, as shown in the GitHub repository.

Genomic prediction (GP) is the procedure whereby the genetic merits of untested candidates are predicted using genome wide marker information. Although numerous examples of GP exist in plants and animals, applications to polyploid organisms are still scarce, partly due to limited genome resources and the complexity of this system. Deep learning (DL) techniques comprise a heterogeneous collection of machine learning algorithms that have excelled at many prediction tasks. A potential advantage of DL for GP over standard linear model methods is that DL can potentially take into account all genetic interactions, including dominance and epistasis, which are expected to be of special relevance in most polyploids. In this study, we evaluated the predictive accuracy of linear and DL techniques in two important small fruits or berries: strawberry and blueberry. The two datasets contained a total of 1,358 allopolyploid strawberry (2n=8x=112) and 1,802 autopolyploid blueberry (2n=4x=48) individuals, genotyped for 9,908 and 73,045 single nucleotide polymorphism (SNP) markers, respectively, and phenotyped for five agronomic traits each. DL depends on numerous parameters that influence performance and optimizing hyperparameter values can be a critical step. Here we show that interactions between hyperparameter combinations should be expected and that the number of convolutional filters and regularization in the first layers can have an important effect on model performance. In terms of genomic prediction, we did not find an advantage of DL over linear model methods, except when the epistasis component was important. Linear Bayesian models were better than convolutional neural networks for the full additive architecture, whereas the opposite was observed under strong epistasis. However, by using a parameterization capable of taking into account these non-linear effects, Bayesian linear models can match or exceed the predictive accuracy of DL. A semiautomatic implementation of the