The Use of Doubled-haploids in Cassava Breeding

Cassava breeding is difficult and, compared with other crops, inefficient. The problems in cassava breeding relate to the length of the breeding cycle, the large genetic load present in the crop, and the heterozygous nature of the parents and progenies evaluated. The production of doubled haploids, through tissue culture techniques, offers interesting advantages. By definition, any process that involves increased homozygosity will result in a decrease of genetic load. Doubled-haploid lines, therefore, are expected to produce better hybrids. Furthermore, the availability of homozygous lines would allow for a gradual and consolidated breeding to improve parental performance in hybrid combinations. This means that the genetic enhancement will benefit from previous gains, like steps in a staircase. The breeder “owns” the genetic superiority of an inbred progenitor but that is not necessarily the case with heterozygous parents. With the introduction of doubled-haploids, the emphasis of cassava breeding shifts from producing large numbers of hybrids (hoping to find a superior one) to improving parents for the production of better hybrids that are designed, not just found. In addition to these advantages, doubled-haploids will facilitate discovery and exploitation of recessive traits and germplasm conservation. One important additional advantage is that germplasm exchange could be greatly facilitated, thus helping to overcome the relative isolation in which many cassava-breeding projects, in different countries, currently operate. INTRODUCTION Cassava improvement has not been as consistent and efficient as in other crops due to many constraints. A typical scheme implies crossing elite clones to produce segregating families (Figure 1). Each individual produced is highly heterozygous. Once a superior genotype is identified (a process that requires about six years), it is vegetatively multiplied to take advantage of the reproductive habits of this crop. This system (except for the vegetative multiplication) has similarities with the ones used for autogamous crops (beans, wheat, rice, etc.) as well as for the hybrid maize industry. However, there is a major difference because cassava is never pushed to produce inbred (homozygous) lines from the segregating progenies of a given cross. The system also bears some similarities with recurrent selection used in allogamous crops (maize), but there is a significant difference because in cassava there is not a clearly defined population whose allelic frequencies are modified through evaluation and selection, as in true recurrent selection schemes. 1. Problems and Limitations in Cassava Breeding From the simplified information provided in Figure 1, it is apparent that a major drawback of cassava breeding is also the length of each breeding cycle. For the reasons below, cassava breeding is slow and inefficient: 1 Cassava Breeding Project, CIAT, Apartado Aereo 67-13, Cali, Colombia. 2 Former cassava breeder at CIAT. Currently at Weaver Popcorn Company, Indiana, USA.