High-Throughput Fluorescent Pollen Tetrad Analysis Using DeepTetrad

Abstract

Meiotic recombination initiates from ~100–200 s of programmed DNA double stranded breaks (DSBs) in plants. Meiotic DSBs can be repaired using homologous chromosomes to generate a crossover. Meiotic crossover is critical for chromosomal segregation and increasing genetic variation. The number of crossovers is limited to one and three per chromosome pair in most plant species. Genetic, epigenetic, and environmental factors control crossover frequency and distribution. Due to the limited number of crossovers it is challenging to measure crossover frequency along chromosomes. We adapted fluorescence-tagged lines (FTLs) that contain quartet1 mutations and linked transgenes expressing dsRed, eYFP, and eCFP in pollen tetrads into the deep learning-based image analysis tool, DeepTetrad. DeepTetrad enables the measurement of crossover frequency and interference by classifying 12 types of tetrads from three-color FTLs in a high-throughput manner, using conventional microscope instruments and a Linux machine. Here, we provide detailed procedures for preparing tetrad samples, tetrad imaging, running DeepTetrad, and analysis of DeepTetrad outputs. DeepTetrad-based measurements of crossover frequency and interference ratio will accelerate the genetic dissection of meiotic crossover control. © 2022, The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature.