Modern phenotyping provided unbiassed computer vision-based and molecular tools for crop improvement. Visualizing previously invisible traits enables the next level of understanding the relationship between genome, environment, and plant performance. Microscopic phenotyping reveals intimate details of the plant-pathogen interactions and helps to discover novel disease resistance genes and mechanisms. However, manual microscopy is very laborious and hardly possible to apply for big phenotyping screens. Microphenomics is a new approach that enables microscopic phenotyping on a large scale. It combines high-throughput automated micro- and macroscopy and Deep learning image analysis tools for achieving the very high throughput and preciseness required for phenomics applications.
BluVision is a Microphenomics framework for high-throughput and precise phenotyping on microscopic and macroscopic levels developed at IPK Leibniz. The system, consisting of hardware, software, and data management components, aims primarily at phenotyping powdery mildew disease of barley and wheat. The phenotypes delivered by the BluVision platform, combined with genomics data, allow the discovery of novel disease-resistant genes of high potential interest to plant pathologists and breeders.
The high level of automatization and throughput of the system allows the phenotyping of large collections of genotypes for genetics and genomics studies. Moreover, the system enables phenotypes that were hardly accessible with manual methods, such as precise quantification of the fungal colonies and finding rare very infection events.
The data, organized according to the FAIR principles, allows the implementation of complete pipelines for complex data analysis like building statistical models, Genome-wide association studies (GWAS), and genomic predictions. Furthermore, the results of the described approaches will be integrated continuously into de.NBI services.
BluVision Micro module
BluVision Micro is an automated high-throughput microscopy system for phenotyping on a cellular and subcellular level. It is based on a Zeiss Axio Scan.Z1 slide scanner designed for high-quality digitalization of microscopy samples, in multiple Z-levels, in bright-field and multi-channel fluorescence. The multidimensional images are analyzed by an automated, artificial intelligence-based pipeline for extracting the relevant phenotypes. Currently, the system is best adapted to phenotyping of interaction of powdery-mildew with cereals, but other microscopic phenotypes are under development or possible by request.
BluVision Macro Module
The core of this module is the Macrobot, a robotized multimodal imaging device developed by IPK Leibniz and Fraunhofer IFF. The Macrobot can acquire images from different samples in a microtiter plate (MTP) containers, for instance, infected leaves, in a highly automated and controlled manner. Thus, the system can precisely and reproducibly quantify the percentage of the infected leaf area with a throughput of up to 10 000 individual samples per day, making it appropriate for phenotyping of large germplasm collections and breeding populations.
Dr. Dimitar Kostadinov Douchkov
Tel.: 039482/ 5285