PERSEUS: an interactive and intuitive web-based tool for pedigree visualization.

SUMMARY: Pedigree-based analyses' prime role is to unravel relationships between individuals in breeding programs and germplasms. This is critical information for decoding the genetics underlying main inherited traits of relevance, and unlocking the genotypic variability of a species to carry out genomic selections and predictions. Despite the great interest, current lineage visualizations become quite limiting in terms of public display, exploration, and tracing of traits up to ancestral donors. PERSEUS is a user-friendly, intuitive, and interactive web-based tool for pedigree visualizations represented as directed graph networks distributed using a force-repulsion method. The visualizations do not only showcase individual relationships among accessions, but also facilitate a seamless search and download of phenotypic traits along the pedigrees. PERSEUS is a promising tool for breeders and scientists, advantageous for evolutionary, genealogy, and diversity analyses among related accessions and species. AVAILABILITY AND IMPLEMENTATION: PERSEUS is freely accessible at https://bioinformatics.cragenomica.es/perseus and GitHub code is available at https://github.com/aranzana-lab/PERSEUS.

LeTra: a leaf tracking workflow based on convolutional neural networks and intersection over union.

BACKGROUND: The study of plant photosynthesis is essential for productivity and yield. Thanks to the development of high-throughput phenotyping (HTP) facilities, based on chlorophyll fluorescence imaging, photosynthetic traits can be measured in a reliable, reproducible and efficient manner. In most state-of-the-art HTP platforms, these traits are automatedly analyzed at individual plant level, but information at leaf level is often restricted by the use of manual annotation. Automated leaf tracking over time is therefore highly desired. Methods for tracking individual leaves are still uncommon, convoluted, or require large datasets. Hence, applications and libraries with different techniques are required. New phenotyping platforms are initiated now more frequently than ever; however, the application of advanced computer vision techniques, such as convolutional neural networks, is still growing at a slow pace. Here, we provide a method for leaf segmentation and tracking through the fine-tuning of Mask R-CNN and intersection over union as a solution for leaf tracking on top-down images of plants. We also provide datasets and code for training and testing on both detection and tracking of individual leaves, aiming to stimulate the community to expand the current methodologies on this topic. RESULTS: We tested the results for detection and segmentation on 523 Arabidopsis thaliana leaves at three different stages of development from which we obtained a mean F-score of 0.956 on detection and 0.844 on segmentation overlap through the intersection over union (IoU). On the tracking side, we tested nine different plants with 191 leaves. A total of 161 leaves were tracked without issues, accounting to a total of 84.29% correct tracking, and a Higher Order Tracking Accuracy (HOTA) of 0.846. In our case study, leaf age and leaf order influenced photosynthetic capacity and photosynthetic response to light treatments. Leaf-dependent photosynthesis varies according to the genetic background. CONCLUSION: The method provided is robust for leaf tracking on top-down images. Although one of the strong components of the method is the low requirement in training data to achieve a good base result (based on fine-tuning), most of the tracking issues found could be solved by expanding the training dataset for the Mask R-CNN model.

GenoDrawing: An Autoencoder Framework for Image Prediction from SNP Markers.

Advancements in genome sequencing have facilitated whole-genome characterization of numerous plant species, providing an abundance of genotypic data for genomic analysis. Genomic selection and neural networks (NNs), particularly deep learning, have been developed to predict complex traits from dense genotypic data. Autoencoders, an NN model to extract features from images in an unsupervised manner, has proven to be useful for plant phenotyping. This study introduces an autoencoder framework, GenoDrawing, for predicting and retrieving apple images from a low-depth single-nucleotide polymorphism (SNP) array, potentially useful in predicting traits that are difficult to define. GenoDrawing demonstrates proficiency in its task using a small dataset of shape-related SNPs. Results indicate that the use of SNPs associated with visual traits has substantial impact on the generated images, consistent with biological interpretation. While using substantial SNPs is crucial, incorporating additional, unrelated SNPs results in performance degradation for simple NN architectures that cannot easily identify the most important inputs. The proposed GenoDrawing method is a practical framework for exploring genomic prediction in fruit tree phenotyping, particularly beneficial for small to medium breeding companies to predict economically substantial heritable traits. Although GenoDrawing has limitations, it sets the groundwork for future research in image prediction from genomic markers. Future studies should focus on using stronger models for image reproduction, SNP information extraction, and dataset balance in terms of phenotypes for more precise outcomes.

An efficient CRISPR-Cas9 enrichment sequencing strategy for characterizing complex and highly duplicated genomic regions. A case study in the Prunus salicina LG3-MYB10 genes cluster.

BACKGROUND: Genome complexity is largely linked to diversification and crop innovation. Examples of regions with duplicated genes with relevant roles in agricultural traits are found in many crops. In both duplicated and non-duplicated genes, much of the variability in agronomic traits is caused by large as well as small and middle scale structural variants (SVs), which highlights the relevance of the identification and characterization of complex variability between genomes for plant breeding. RESULTS: Here we improve and demonstrate the use of CRISPR-Cas9 enrichment combined with long-read sequencing technology to resolve the MYB10 region in the linkage group 3 (LG3) of Japanese plum (Prunus salicina). This region, which has a length from 90 to 271 kb according to the P. salicina genomes available, is associated with fruit color variability in Prunus species. We demonstrate the high complexity of this region, with homology levels between Japanese plum varieties comparable to those between Prunus species. We cleaved MYB10 genes in five plum varieties using the Cas9 enzyme guided by a pool of crRNAs. The barcoded fragments were then pooled and sequenced in a single MinION Oxford Nanopore Technologies (ONT) run, yielding 194 Mb of sequence. The enrichment was confirmed by aligning the long reads to the plum reference genomes, with a mean read on-target value of 4.5% and a depth per sample of 11.9x. From the alignment, 3261 SNPs and 287 SVs were called and phased. A de novo assembly was constructed for each variety, which also allowed detection, at the haplotype level, of the variability in this region. CONCLUSIONS: CRISPR-Cas9 enrichment is a versatile and powerful tool for long-read targeted sequencing even on highly duplicated and/or polymorphic genomic regions, being especially useful when a reference genome is not available. Potential uses of this methodology as well as its limitations are further discussed.

Characterization of Japanese Plum (Prunus salicina) PsMYB10 Alleles Reveals Structural Variation and Polymorphisms Correlating With Fruit Skin Color.

The red to blue hue of plant organs is caused due to anthocyanins, which are water-soluble flavonoid pigments. The accumulation of these pigments is regulated by a complex of R2R3-MYB transcription factors (TFs), basic-helix-loop-helix (bHLH), and WD-repeat (WDR) proteins (MBW complex). In Rosaceae species, R2R3-MYBs, particularly MYB10 genes, are responsible for part of the natural variation in anthocyanin colors. Japanese plum cultivars, which are hybrids of Prunus salicina, have high variability in the color hue and pattern, going from yellow-green to red and purple-blue, probably as a result of the interspecific hybridization origin of the crop. Because of such variability, Japanese plum can be considered as an excellent model to study the color determination in Rosaceae fruit tree species. Here, we cloned and characterized the alleles of the PsMYB10 genes in the linkage group LG3 region where quantitative trait loci (QTLs) for the organ color have been mapped to other Prunus species. Allele segregation in biparental populations as well as in a panel of varieties, combined with the whole-genome sequence of two varieties with contrasting fruit color, allowed the organization of the MYB10 alleles into haplotypes. With the help of this strategy, alleles were assigned to genes and at least three copies of PsMYB10.1 were identified in some varieties. In total, we observed six haplotypes, which were able to characterize 91.36% of the cultivars. In addition, two alleles of PsMYB10.1 were found to be highly associated with anthocyanin and anthocyanin-less skin. Their expression during the fruit development confirms their role in the fruit skin coloration. Here, we provide a highly efficient molecular marker for the early selection of colored or non-colored fruits in Japanese plum breeding programs.

CoExp: A Web Tool for the Exploitation of Co-expression Networks.

Gene co-expression networks are a powerful type of analysis to construct gene groupings based on transcriptomic profiling. Co-expression networks make it possible to discover modules of genes whose mRNA levels are highly correlated across samples. Subsequent annotation of modules often reveals biological functions and/or evidence of cellular specificity for cell types implicated in the tissue being studied. There are multiple ways to perform such analyses with weighted gene co-expression network analysis (WGCNA) amongst one of the most widely used R packages. While managing a few network models can be done manually, it is often more advantageous to study a wider set of models derived from multiple independently generated transcriptomic data sets (e.g., multiple networks built from many transcriptomic sources). However, there is no software tool available that allows this to be easily achieved. Furthermore, the visual nature of co-expression networks in combination with the coding skills required to explore networks, makes the construction of a web-based platform for their management highly desirable. Here, we present the CoExp Web application, a user-friendly online tool that allows the exploitation of the full collection of 109 co-expression networks provided by the CoExpNets suite of R packages. We describe the usage of CoExp, including its contents and the functionality available through the family of CoExpNets packages. All the tools presented, including the web front- and back-ends are available for the research community so any research group can build its own suite of networks and make them accessible through their own CoExp Web application. Therefore, this paper is of interest to both researchers wishing to annotate their genes of interest across different brain network models and specialists interested in the creation of GCNs looking for a tool to appropriately manage, use, publish, and share their networks in a consistent and productive manner.