Computational Approaches For Analysis Of High-throughput Genotyping And Phenotyping Data In Plants

High Throughput technologies in the last decade have enabled availability of components of "omics" approaches that contain genomics, proteomics, transcriptomics, epigenomics, and metabolomics. Integrated "omics" approaches have more potential in aiding crop breeding, leading to a new approach-"phenomics"-involving high-throughput analysis of physical and biochemical traits of an organism. Genomics and phenomics are two fundamentally important branches of biological sciences and they stand at both ends of the multiple "omics" families. A central goal of current biology is to establish complete functional links between the genome and phenome, the so-called genotype-phenotype map.Living cells are the product of gene expression programs that involve the regulated transcription of thousands of genes. The elucidation of transcriptional regulatory networks is thus needed to understand how the cell’s working mechanism and can be useful for the discovery of novel therapeutic targets. In genomics, prediction of gene regularity network (GRN) from expression data is a challenging task. There are many methods that have been developed to address this challenge ranging from supervised to unsupervised methods. Most promising methods are based on support vector machine (SVM).There is a need for comprehensive analysis of the prediction accuracy of supervised method SVM using different kernels on different biological experimental conditions and network size. We developed a tool (CompareSVM) based on SVM to compare different kernel methods for inference of GRN. Using CompareSVM, we investigated and evaluated the different SVM kernels methods on simulated data sets of microarray and next generation sequencing of different sizes in detail. The results obtained from CompareSVM showed that accuracy of inference method depends on upon the nature of experimental condition and size of the network.Constraints in plant phenotyping capability limit our ability to dissect the genetics of quantitative traits, especially those related to yield and stress tolerance (e.g., yield potential, increased drought, heat tolerance, and nutrient efficiency, etc.). The development of effective high-throughput phenotyping platforms remains a bottleneck. However, progress in aeronautics, sensors, and high-performance computing are paving the way. High throughput phenotyping is an important technology to dissect phenotypic components in plants. In order to quantify plant growth and performance based on phenotypic traits, efficient image processing, and feature extraction is essential. There is a need for systems that can support image data transfer from different acquisition environments and large-scale image analysis for different plant species based on real time imaging data gathered from different spectra. High-throughput phenotyping platforms have been developed that enables to capture extensive and intensive phenotype data from plants in a non-destructive manner. Accordingly, these developments advance our view on plant growth and performance with responses to the changing climate and environment. Therefore increasing the efficiency of crop genetic improvement to meet the needs of future generations. Knowledge of recent development in high-throughput plant phenotyping infrastructures based on imaging techniques and corresponding principles for phenotype data analysis has been discussed.The analysis of digital images is one of the important task in plant phenotyping to evaluate plant parameters in a non-invasive fashion. A range of different screening systems with varying requirements for the image analysis have been developed and are in part commercially available. Segmentation and identification of plant organs, the especially individual leaf are one of the biggest challenges in image-based plant phenotyping systems. Fully Automated Systems for phenotyping provide a constant environment for image acquisition but result in very high labor, cost, and maintenance. There is a need for a more flexible system that can adapt to varying plant background, fluctuating illumination and similar changing problems. A freely available ImageJ plugin, HTPPA, has been developed by extending the state of art image processing algorithms libraries of ImageJ, along with some other useful methods to explore high throughout phenotyping in a comprehensive way. By utilizing plant structural and morphological properties to improve plant and individual leaf segmentation, we can facilitate large-scale high throughput phenotyping to establish a relationship between genotype and phenotype.Technological advancement has increased the breadth of available omics data, from whole genome sequencing data to extensive transcriptomics, methylomic, and metabolomic data. A key goal is to establish complete function link between the genotype and phenotype, by identification of effective models that predict phenotypic traits and outcome. High throughput and high dimensional genotyping and phenotyping enable to discover gray areas between genotypes and phenotypes using the principles of genome-wide association studies (GWAS). Application of GWAS and analogous methodologies and incorporation of multiple omics data begin to discover the contribution of genetic variation to phenotypic diversity. Integrating "omics" data at broad levels by using the systems-biology approach is paramount to further bridging the gaps between genomics and phenomics and eventually making accurate predictions of phenotypes based on genetic contribution.