Search

The parasitic angiosperms,Striga hermonthicaandStriga asiatica, severely constrain cereal production in sub-Saharan Africa by causing huge losses in grain yield. Understanding the diversity ofStrigapopulations is important because it allows identification of races or biotypes thus improving chances of breeding success. Amplified fragment length polymorphism (AFLP) analysis was used to study genetic diversity among 17 populations ofS. asiaticaand 24 populations ofS. hermonthicafrom Kenya. A total of 349 DNA fragments ranging from 51 to 500 bp were obtained from fourEcoRI andMseI primer combinations. Genetic distances forS. asiaticapopulations ranged from 0.009 to 0.116 with a mean of 0.032.S. hermonthicapopulations had a genetic distance that ranged from 0.007 to 0.025 with a mean of 0.015. Only two clusters were found inS. asiaticapopulations whereas no apparent structure was evident inS. hermonthicapopulations. There was no evidence of isolation by distance for the two species. Although the low genetic diversity suggestsStrigais relatively uniform across the populations studied, it is possible that pathogenicity and virulence genes may be located in genomic regions that were not sampled. The data, however, does not provide evidence to support diversification of bothStrigaspecies in the region where the study was conducted. [ABSTRACT FROM AUTHOR]

In this study, we collected and analyzed DNA sequence data for 789 previously mapped RFLP probes from Sorghum bicolor (L.) Moench. DNA sequences, comprising 894 non-redundant contigs and end sequences, were searched against three GenBank databases, nucleotide (nt), protein (nr) and EST (dbEST), using BLAST algorithms. Matching ESTs were also searched against nt and nr. Translated DNA sequences were then searched against the conserved domain database (CDD) to determine if functional domains/motifs were congruent with the proteins identified in previous searches. More than half (500/894 or 56%) of the query sequences had significant matches in at least one of the GenBank searches. Overall, proteins identified for 148 sequences (17%) were consistent among all searches, of which 66 sequences (7%) contained congruent coding domains. The RFLP probe sequences were also evaluated for the presence of simple sequence repeats (SSRs) and 60 SSRs were developed and assayed in an array of sorghum germplasm comprising inbreds, landraces and wild relatives. Overall, these SSR loci had lower levels of polymorphism (D = 0.46, averaged over 51 polymorphic loci) compared with sorghum SSRs that were isolated by library hybridization screens (D = 0.69, averaged over 38 polymorphic loci). This result was probably due to the relatively small proportion of di-nucleotide repeat-containing markers (42% of the total SSR loci) obtained from the DNA sequence data. These di-nucleotide markers also contained shorter repeat motifs than those isolated from genomic libraries. Based on BLAST results, 24 SSRs (40%) were located within, or near, previously annotated or hypothetical genes. We determined the location of 19 of these SSRs relative to putative coding regions. In general, SSRs located in coding regions were less polymorphic (D = 0.07, averaged over three loci) than those from gene flanking regions, UTRs and introns (D = 0.49, averaged over 16 loci). The sequence information and SSR loci generated through this study will be valuable for application to sorghum genetics and improvement, including gene discovery, marker-assisted selection, diversity and pedigree analyses, comparative mapping and evolutionary genetic studies. [ABSTRACT FROM AUTHOR]