Linkage & Selective Mapping and QTL Analysis in a Harosoy x Clark population of 300 RILs

Four of the 12 major Glycine max ancestors of all modern elite U.S.A. soybean cultivars were the grandparents of Harosoy and Clark, so a Harosoy x Clark population would include some of that genetic diversity. A mating of eight Harosoy and eight Clark plants generated eight F1 plants. The eight F1:2 families were advanced via a plant-to-row selfing method to produce 300 F6-derived RILs that were genotyped with 266 SSR, 481 SNP, and 4 classical markers. SNPs were genotyped with the Illumina 1536-SNP assay. Three linkage maps, SSR, SNP, and SSR-SNP, were constructed with a genotyping error of < 1 %. Each map was compared with the published soybean consensus map. The best subset of 94 RILs for a high-resolution framework (joint) map was selected based on the expected bin length statistic computed with MapPop. The QTLs of seven traits measured in a 2-year replicated performance trial of the 300 RILs were identified using composite interval mapping (CIM) and multiple-interval mapping (MIM). QTL x Year effects in multiple trait analysis were compared with results of multiple-interval mapping. QTL x QTL effects were identified in MIM.

An Extended Kernel for Generalized Multiple-Instance Learning

The multiple-instance learning (MIL) model has been successful in areas such as drug discovery and content-based image-retrieval. Recently, this model was generalized and a corresponding kernel was introduced to learn generalized MIL concepts with a support vector machine. While this kernel enjoyed empirical success, it has limitations in its representation. We extend this kernel by enriching its representation and empirically evaluate our new kernel on data from content-based image retrieval, biological sequence analysis, and drug discovery. We found that our new kernel generalized noticeably better than the old one in content-based image retrieval and biological sequence analysis and was slightly better or even with the old kernel in the other applications, showing that an SVM using this kernel does not overfit despite its richer representation.

Anxiolytic-Like Property of Risperidone and Olanzapine as Examined in Multiple Measures of Fear in Rats

Atypical antipsychotics are also used in the treatment of anxiety-related disorders. Clinical and preclinical evidence regarding their intrinsic anxiolytic efficacy has been mixed. In this study, we examined the potential anxiolytic-like effects of risperidone and olanzapine, and compared them with haloperidol, chlordiazepoxide (a prototype of sedative–anxiolytic drug) or citalopram (a selective serotonin reuptake inhibitor). We used a composite of two-way avoidance conditioning and acoustic startle reflex model and examined the effects of drug treatments during the acquisition phase (Experiment 1) or extinction phase (Experiments 2 and 3) on multiple measures of conditioned and unconditioned fear/anxiety-like responses. In Experiment 4, we further compared risperidone, olanzapine, haloperidol, citalopram and chlordiazepoxide in a standard elevated plus maze test. Results revealed three distinct anxiolytic-like profiles associated with risperidone, olanzapine and chlordiazepoxide. Risperidone, especially at 1.0 mg/kg, significantly decreased the number of avoidance responses, 22 kHz ultrasonic vocalization, avoidance conditioning-induced hyperthermia and startle reactivity, but did not affect defecations or time spent on the open arms. Olanzapine (2.0 mg/kg, sc) significantly decreased the number of avoidance responses, 22 kHz vocalization and amount of defecations, but it did not inhibit startle reactivity and time spent on the open arms. Chlordiazepoxide (10 mg/kg, ip) significantly decreased the number of 22 kHz vocalization, avoidance conditioning-induced hyperthermia and amount of defecations, and increased time spent on the open arms, but did not decrease avoidance responses or startle reactivity. Haloperidol and citalopram did not display any anxiolytic-like property in these tests. The results highlight the importance of using multiple measures of fear-related responses to delineate behavioral profiles of psychotherapeutic drugs.

QUANTITATIVE TRAIT LOCI FOR AGRONOMIC AND END-USE QUALITY PERFORMANCE AND THE EFFECT OF SOILBORNE WHEAT MOSAIC VIRUS IN A HARD WINTER WHEAT POPULATION IN NEBRASKA

To better understand agronomic and end-use quality in wheat (Triticum aestivum L.) we developed a population containing 154 F6:8 recombinant inbred lines (RILs) from the cross TAM107-R7/Arlin. The parental lines and RILs were phenotyped at six environments in Nebraska and differed for resistance to Wheat soilborne mosaic virus (WSBMV), morphological, agronomic, and end-use quality traits. Additionally, a 2300 cM genome-wide linkage map was created for quantitative trait loci (QTL) analysis. Based on our results across multiple environments, the best RILs could be used for cultivar improvement. The population and marker data are publicly available for interested researchers for future research. The population was used to determine the effect of WSBMV on agronomic and end-use quality and for the mapping of a resistance locus. Results from two infected environments showed that all but two agronomic traits were significantly affected by the disease. Specifically, the disease reduced grain yield by 30% of susceptible RILs and they flowered 5 d later and were 11 cm shorter. End-use quality traits were not negatively affected but flour protein content was increased in susceptible RILs. The resistance locus SbmTmr1 mapped to 27.1 cM near marker wPt-5870 on chromosome 5DL using ELISA data. Finally, we investigated how WSBMV affected QTL detection in the population. QTLs were mapped at two WSBMV infected environments, four uninfected environments, and in the resistant and susceptible RIL subpopulations in the infected environments. Fifty-two significant (LOD≥3) QTLs were mapped in RILs at uninfected environments. Many of the QTLs were pleiotropic or closely linked at 6 chromosomal regions. Forty-seven QTLs were mapped in RILs at WSBMV infected environments. Comparisons between uninfected and infected environments identified 20 common QTLs and 21 environmentally specific QTLs. Finally, 24 QTLs were determined to be affected by WSBMV by comparing the subpopulations in QTL analyses within the same environment. The comparisons were statistically validated using marker by disease interactions. These results showed that QTLs can be affected by WSBMV and careful interpretation of QTL results is needed where biotic stresses are present. Finally, beneficial QTLs not affected by WSBMV or the environment are candidates for marker-assisted selection.

Evaluating Measurement Invariance with Censored Ordinal Data: A Monte Carlo Comparison of Alternative Model Estimators and Scales of Measurement

Evaluations of measurement invariance provide essential construct validity evidence. However, the quality of such evidence is partly dependent upon the validity of the resulting statistical conclusions. The presence of Type I or Type II errors can render measurement invariance conclusions meaningless. The purpose of this study was to determine the effects of categorization and censoring on the behavior of the chi-square/likelihood ratio test statistic and two alternative fit indices (CFI and RMSEA) under the context of evaluating measurement invariance. Monte Carlo simulation was used to examine Type I error and power rates for the (a) overall test statistic/fit indices, and (b) change in test statistic/fit indices. Data were generated according to a multiple-group single-factor CFA model across 40 conditions that varied by sample size, strength of item factor loadings, and categorization thresholds. Seven different combinations of model estimators (ML, Yuan-Bentler scaled ML, and WLSMV) and specified measurement scales (continuous, censored, and categorical) were used to analyze each of the simulation conditions. As hypothesized, non-normality increased Type I error rates for the continuous scale of measurement and did not affect error rates for the categorical scale of measurement. Maximum likelihood estimation combined with a categorical scale of measurement resulted in more correct statistical conclusions than the other analysis combinations. For the continuous and censored scales of measurement, the Yuan-Bentler scaled ML resulted in more correct conclusions than normal-theory ML. The censored measurement scale did not offer any advantages over the continuous measurement scale. Comparing across fit statistics and indices, the chi-square-based test statistics were preferred over the alternative fit indices, and ΔRMSEA was preferred over ΔCFI. Results from this study should be used to inform the modeling decisions of applied researchers. However, no single analysis combination can be recommended for all situations. Therefore, it is essential that researchers consider the context and purpose of their analyses.