Gene segregation in autotetraploids: Prediction from meiotic configurations

Current systems of estimating gene segregation in autotetraploids usually are based on two extreme methods: chromosome segregation with absolute centromere linkage, or maximum equational segregation with crossing over always occurring between the centromeres and marker genes. The hiatus between the two methods is unacceptable quantitatively, and segregation of the d gene in tomato shows that neither method predicts an acceptable fit. We present a new meiotic configuration (MC) method of analyzing tetrasomic inheritance in autotetraploids that allows more accurate estimations of all intermediate types of segregation. The method is based on a maximum of two chiasmata per bivalent and four per quadrivalent. The theoretically expected numbers of bivalents and chain and circle quadrivalents are derived first. Chromosome frequencies from these configurations are then used to determine relative contributions from each configuration to the gamete genotypes by way of newly developed tables for gene segregation in various autotetraploid genotypes. The large sample sizes from classic 4x tomato (Solanum lycopersicum L.) data are used to demonstrate the stepwise method of calculating gamete genotypes, and the method gives acceptable fits for the data tested. The interspecific tetraploid hybrid Solanum lycopersicum x S. pimpinellifolium had good fits to expected autotetraploid segregation, supporting its autoploid nature and the fact that taxonomic rank should not be a consideration in designating categories of polyploids. Autotetraploid allozyme segregation data for several genes in one genus of another family had acceptable and usually better fits to the MC expectations than to the standard methods.