Single clone discovery using an image-based cell isolation platform

Abstract Characterizing both functional and genetic heterogeneity among a pool of cells remains a major scientific challenge in immunology, cancer research, neurobiology and developmental biology. Isolating cells with the same phenotype is key to understand such heterogeneity. We would like to develop a non-fluidic, unique technology, LCD aided selection under microscope (LASUM, Fig.1) to address this need. During the phase I of this project, we completed the alpha version of prototype design and demonstrated its utility in three proof-of-concept experiments: (a) blood cell removal; (b) phage enrichment on single beads; (c) isolation of tumor-killing immune cell clones. We would like to continue our effort to bring this technology to market, which will be low-cost, high throughput, debris resistant, image-based with operation simplicity as preparing a microscope slide or washing a microtiter plate. Besides the device, we will unify the fragmentated cell isolation market using two associated kits: Enrich-Live and Enrich- Seq. The successful commercialization of this device/kits will enable thousands of cell- biology labs and hospitals in manipulating single cells on a computer screen.Narrative We here propose the continued development of a cost-effective and biocompatible single cell isolation technology that utilizes the Liquid Crystal Display (LCD) with growing resolution to create a dynamic selection photomask (Fig.1), while the identity and classification of cells and cell-pairs are achieved by pattern-recognition algorithms. Cell isolation can be achieved in two ways: unwanted particle populations will be trapped by local photo-polymerization /gelation, the selected cells can be collected by elution; alternatively, cells of interest can be trapped in place, while unwanted cells/debris can be washed away. The method can be used to select and isolate cells based on organelle translocation, chromatin morphology, phagocytosis, synapse formation, protein co-localization, cell-cell interactions and other time-resolved cellular behaviors which cannot be achieved by current sorting platforms in a high-throughput manner. To make it commercially viable (andlt;$10K production cost), several technical goals are critical: the proprietary error-tolerant algorithm of particle locating, the use of mass-produced components and the software portal for including external high-resolution images from third party platforms. To make the throughput comparable to common micro-fluidic sorters, optimized image pattern recognition algorithms and parallel computation tools will be implemented. Multiple applications will be used as test cases for the validation of device/kits during Phase II development.