DeepTetrad: high-throughput analysis of meiotic tetrads by deep learning in plants

Abstract

15Meiotic crossovers facilitate chromosome segregation and create new combinations of alleles 16 in gametes. Crossover frequency varies along chromosomes and crossover interference limits 17 the coincidence of closely spaced crossovers. Crossovers can be measured by observing the 18 inheritance of linked transgenes expressing different colors of fluorescent protein in 19 Arabidopsis pollen tetrads. Here we establish DeepTetrad, a deep learning-based image 20 recognition package for pollen tetrad analysis that enables high-throughput measurements of 21 crossover frequency and interference in individual plants. DeepTetrad will accelerate genetic 22 dissection of mechanisms that control meiotic recombination. 23 24 Main 25 Meiosis consists of two consecutive nuclear divisions and produces four haploid gametes from 26 a single diploid cell in sexually reproducing eukaryotes 1 . In Arabidopsis male meiosis, ~200-27 DeepTetrad 2 250 meiotic DNA double-strand breaks (DSBs) are induced in the genome by a DNA 28 topoisomerase VI-like complex to initiate meiotic recombination 2-4 . Of these DSBs, only ~8-29 11 are repaired as crossovers (COs) using a homologous chromosome (homolog). Thus, male 30 meiosis in the Arabidopsis genome, which comprises five chromosomes, results in an average 31 of ~1.8 crossovers between homologs. This low number suggests the existence of 32 mechanisms that limit crossovers, a phenomenon that is observed in most eukaryotes 2 .33 Meiotic DSB and CO frequencies are controlled by genetic and epigenetic factors and are 34 non-randomly distributed along chromosomes, with higher levels around gene promoters and 35 terminators and lower levels across the centromeres 5-7 . 36 At least two pathways (Type I and Type II), contribute to CO formation 2,3 . The Type I pathway 37 leads to interfering COs that prevent the coincident occurrence of closely spaced CO on the 38 same pair of chromosomes 2,8,9 . In plants, interfering COs represent ~80-85% of total COs and 39 are dependent on the ZMM proteins (ZIP4, MSH4, MSH5, MER3, HEI10, SHOC1, PTD, MLH1, 40 MLH3). The remaining ~10-15% of COs are non-interfering and occur via the Type II 41 pathway 10 . Non-interfering COs are resolved by the MUS81 endonuclease and are restricted 42 by anti-recombination factors such as FANCM, RECQ4A, RECQ4B, and FIGL1 11-15 . 43 Disruption of anti-recombination factors can increase the number of Type II COs in plants, 44 which has the potential to create new combinations of desirable alleles that can improve crop 45 varieties 13,16 . Therefore, high-throughput detection and understanding of CO frequency and 46 interference have important implications for our understanding of the control of meiotic 47 QRT1 gene encoding a pectin methylestrase results in the four pollen products of male 51 meiosis remaining attached to one another, allowing classical tetrad analysis. Each FTL has 52 a transgenes that expresses eYFP (Y), dsRed (R) or eCFP (C) fluorescent proteins in mature 53 pollen using the post-meiotic LAT52 promoter. Gen...