Soybean (Glycine max (L.) Merrill) is the most important leguminous crop in the world due to the high contents of protein and oil, and accumulation of various physiologically active substances. However, most of the genomic and economic traits in soybean are quantitative, controlled by multiple genes and easily affected by environmental conditions. Based on recombinant inbred lines (F 8 ), a genetic linkage map with 177 RFLP, 150 SSR, 28 AFLP markers and 5 phenotypic markers was constructed. The map covered a distance of 2663.6 cM of the soybean genome comprising 20 linkage groups. The average distance between two adjacent marker loci was 7.89 cM. In this population, we detected thirty-nine QTLs for all the reproductive development and seed quality traits investigated, that is, three for flowering time (FT1-3), four for maturity (HAV1-4), three for reproductive period (RP1-3), three for seed hardness (RAS1-3), five for viability of seed (VIS1-5), four for germination rate of seed (GRS1-4), five for water absorbability of seed (WAS1-5) and twelve QTLs for seed weight (SWE1-6 and SWH1-6). Out of these QTLs, twenty-eight were detected in nearly the same regions of the linkage map by both IM (interval mapping) and CIM (composite interval mapping) analysis. The proportion of variance explained of these QTLs ranged from 3.4 % to 67.1 %. Epistatic interactions were detected among various QTLs. Especially there was a strong interaction among the effects of FT1 and FT2, and FT1 and FT3. Multiple correlation coefficients between FT1 and FT2, and FT1 and FT3 accounted for 79.6 % and 74.1 % of the phenotypic variation of flowering time, respectively.

Improvement of the quality and quantity of soybean seed constituents is one of the most important objectives in soybean breeding. Although the quality of seed constituents has been studied extensively, information on the quantity is still limited. In order to analyze the genetic basis of these traits, recombinant inbred lines (RILs) derived from a cross between Glycine max (L.) Merrill variety Misuzudaizu and variety Moshidou Gong 503 were planted in two environments and evaluated for seed protein and lipid contents. Protein and lipid contents were determined by NIR transmittance spectroscopy using an intact single seed. The broad sense heritability of the traits ranged from 0.73 to 0.79 in our RIL population. Single-factor analysis of variance, interval mapping and composite interval mapping were used to detect significant associations between the traits and genetic markers. A total of 17 QTLs, 10 for proteins and 7 for lipids, which were significant in at least one environment were identified. Each QTL explained the total phenotypic variation for protein and lipid contents in the range from 3.4 % to 29.7 % and 6.1 % to 10.1 %, respectively. Among all the detected QTLs, three for the protein content and three for the lipid content were detected in both environments. The negative correlation between protein and lipid contents was also confirmed. Epistatic interactions were detected in this study, as another source of genetic variation in our population. The results obtained in our study may serve as a base for analyzing the genetic control of protein and lipid contents and may eventually enable to change the seed constituents.

The molecular types and genetic heterogeneity of Cryptococcus neoformans and C. gattii clinical isolates in Malaysia were determined in this study. Of 44 C. neoformans collected between 1980 and 2003, 42 (95.5%) were molecular type VNI, 2 (4.5%) were molecular type VNII. Of 17 C.gattii isolates, 13 (76.5%) were molecular type VGI, and 4 (23.5%) were molecular type VGII. A difference was noted when comparing the molecular types of cryptococcal isolates in the earlier and recent cases of cryptococcosis. While both molecular types VNI and VGI were equally predominant in the earlier cases of cryptococcosis, VNI was the most predominant molecular type isolated from the recent cases. VNII was a new molecular type, isolated from 5.1% of the recent cases. All the bird dropping isolates were molecular type VNI. The genetic heterogeneity of the two predominant molecular types, i.e., VNI, VGI clinical isolates and bird dropping isolates of C. neoformans were further determined by polymerase chain reaction (PCR) fingerprinting method, using (GTG)5 as single primer. Two clusters of cryptococcal isolates were distinguished at 68.5% of similarity, with cluster I consisting of VNI isolates and cluster II consisting of VGI isolates. Each cluster was further subdivided into three subtypes at >/=80% of similarity. Fourteen bird dropping isolates were grouped into a subtype within VN1, sharing 82.7% of similarity with the clinical isolates. A higher degree of similarities, ranging from 93.4-97.6% was noted between 3 bird dropping isolates with the clinical isolates in another subtype. This study demonstrated the existence of various molecular types of C. neoformans isolates in Malaysia and the genetic heterogeneity within the predominant molecular types. The study also provides evidence for genetic relatedness of clinical isolates with bird dropping isolates in the environment.