Spearman’s law of diminishing returns (SLODR) posits that at higher levels of general cognitive ability, the general factor (g) performs less well in explaining individual differences in cognitive test performance. The present study used factor mixture modeling to investigate SLODR in the Kaufman Assessment Battery for Children--Second Edition (KABC-II). Factor mixture modeling was a useful method to study SLODR because group membership was determined based on probabilities derived from the model. A second-order confirmatory factor model, consistent with three-stratum theory (Carroll, 1993), was modeled as a within-class factor structure. The fit of several models with varying number of classes and factorial invariance restrictions were compared. A sex covariate was also included with the models that provided the best fit for the data. The results indicated that a two-class model, which allowed for g mean differences, and class-specific g variances and subtest residual variances, provided the most parsimonious explanation of the data. Consistent with SLODR, the second-order general factor explained less subtest variance and less variance in the first-order factors for those of higher general ability. The standardized subtest residual variances were also larger in the high ability class than in the low ability class. Controlling for g, boys performed higher than girls in visual-spatial ability in each of the low and high ability classes. The findings from this study have implications for future research on the interpretation of intelligence test scores across the ability distribution. / text

Advancements in microelectronic devices and computational and storage technologies enable the collection of high volume, high speed and high dimension data in many applications. Due to the high dimensionality of these measurements, exact dependence of the observations on the various parameters or variables may not be exactly known. Factor analysis (FA) is a useful multivariate technique to exploit the redundancies among observations and reveal their dependence to some latent variables called factors. Some major issues of the conventional FA are high arithmetic complexity for real-time online implementation, assumption of static system parameters, the demand of interval forecasting, robustness against outlying observations and model selection in problems with high dimension but low number of samples (HDLS). This thesis addresses these issues and proposes new extensions to the existing FA algorithms. First, in order to reduce the arithmetic complexity, we propose new recursive FA algorithms (RFA) that recursively compute only the dominant Principal Components (PCs) and eigenvalues in the major subspace tracked by efficient subspace tracking algorithms. Specifically, two new approaches are proposed for updating the PCs and eigenvalues in the classical fault detection problem with different tradeoff between accuracy and arithmetic complexity, namely rank-1 modification and deflation. They significantly reduce the online arithmetic complexity and allow the adaption to time-varying system parameters. Second, we extend the RFA algorithm to forecasting of time series and propose a new recursive dynamic factor analysis (RDFA) algorithm for electricity price forecasting. While the PCs are recursively tracked by the subspace algorithm, a random walk or a state dynamical model can be incorporated to describe the latest state of the time-varying auto-regressive (AR) model built from the factors. This formulation can be solved by the celebrated Kalman filter (KF), which in turn allows future values to be forecasted with estimated confidence intervals. Third, we propose new robust covariance and outlier detection criteria to improve the robustness of the proposed RFA and RDFA algorithms against outlying observations based on the concept of robust M-estimation. Experimental results show that the proposed methods can effectively suppress the adverse contributions of the outliers on the factors and PCs. Finally, in order to improve the consistency of model selection and facilitate the estimation of p-values in HDLS problems, we propose a new automatic model selection method based on ridge partial least squares and recursive feature elimination. Furthermore, a novel performance criterion is proposed for ranking variables according to their consistency of being chosen in different perturbation of the samples. Using this criterion, the associated p-values can be estimated under the HDLS setting. Experimental results using real gene cancer microarray datasets show that improved prognosis can be obtained by the proposed approach as compared with conventional techniques. Furthermore, to quantify their statistical significance, the p-value of the identified genes are estimated and functional analysis of the significant genes found in the diffused large B-cell lymphoma (DLBCL) gene microarray data is performed to validate the findings. While we focus in a few engineering problems, these algorithms are also applicable to other related applications. / published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

No description available.

During embryogenesis, hematopoietic cells arise from the lateral plate mesoderm (LPM) following gastrulation. The transcriptional program required for this LPM to blood switch is not fully understood. Previous work on a zebrafish mutant with a deletion in the cdx4 gene demonstrated the importance of this caudal transcription factor in the LPM to blood transition. To explain how cdx4 regulates embryonic hematopoiesis, two main approaches were taken in this thesis. The first part of the thesis describes a chemical genetics screen that identified cdx4 interacting pathways. To find small molecules that could rescue the loss of red blood cells caused by the cdx4 deletion, cdx4 mutant embryos were incubated with 2640 compounds from the beginning of the gastrula stage to the 10-somite stage. Two related psoralen compounds, Bergapten (Ber) and 8-methoxypsoralen (8-MOP), rescued the erythroid progenitors in the cdx4 mutants. This rescue is closely linked to the compounds' effects on anteriorposterior patterning, reminiscent of retinoic acid pathway compounds. The second part of my thesis identifies a Cdx4-Sall4 transcriptional module in the LPM. Chromatin-immunoprecipitation coupled to sequencing (ChIP-seq) and microarray analysis revealed that Cdx4 directly regulates cdx4 and a zinc finger transcription factor spalt-like 4 (sall4) transcription. Sall4 ChIP-seq showed that Sall4 also binds to its own locus and to the cdx4 locus, suggesting an auto- and cross-regulation between two transcription factors. In addition, Cdx4 and Sall4 bind to common genomic regions proximal to mesodermal progenitor (tbx16 and mespa) and hematopoietic genes (scl, gata2a, and ldb1a), indicating Cdx4 and Sall4 co-regulate key genes that are required for LPM and blood specification. sall4 knockdown in the cdx4 mutants demonstrated that Sall4 synergizes with Cdx4 in regulating embryonic hematopoiesis. These findings suggest that auto- and cross-regulation of Cdx4 and Sall4 establish a stable circuit in the LPM that facilitates the activation of blood-specific program as development proceeds. How undifferentiated germ layers transition into various tissues is a key question in developmental biology. My thesis establishes a model based on LPM to blood transition, which is also applicable to other studies on germ layer specification.

In plants, flowering time is a tightly regulated process where several environmental and endogenous cues fine-tune the time of flowering. In Arabidopsis, four major genetic pathways regulate flowering time, namely photoperiod, vernalization, autonomous, and phytohormone gibberellic acid (GA) pathways. Arabidopsis is a facultative long day (LD) plant. LD promotes flowering whereas flowering is delayed in short day (SD) conditions. Here, we identified a basic-helix-loop-helix (bHLH) transcription factor called bHLH93 that is necessary to promote flowering only in SD. Also, photoperiod plays more critical roles in regulation of flowering time of bhlh93 mutant compared to GA and vernalization pathways. Thus, bHLH93 might represent a novel transcription factor absolutely required for Arabidopsis thaliana to evolve as a facultative LD plant. bhlh93 mutants also show severe adult phenotype such as shorter stature, curly and darker green leaves, and reduced fertility compared to wild type plants. These results suggest that bHLH93 controls plant stature, fertility and chlorophyll content in Arabidopsis. bHLH93 is expressed in a tissue-specific and developmental stage-dependent manner. bHLH93-YFP protein is localized in the nucleus. bHLH93 homodimerizes in yeast, and it has strong transcription activation activity in yeast. These data suggest that, like other bHLH proteins, bHLH93 may function as a transcriptional regulator in the nucleus controlling gene expression. We have identified floral repressor MAF5 as a major target of bHLH93 to promote flowering in SD. bHLH93 binds to MAF5 promoter element in vivo and in vitro. Other than MAF5, FLC and MAF1-2 are also up-regulated in bhlh93 but at a lower level than MAF5. The activation of multiple floral repressors correlates with bhlh93 flowering phenotype. Taken together, these data suggest that bHLH93 may provide selective advantage for evolution of facultative flowering behavior under varying environmental conditions for reproductive success. / text

published_or_final_version / Pharmacology / Master / Master of Philosophy

published_or_final_version / Paediatrics / Master / Master of Philosophy

published_or_final_version / Biochemistry / Doctoral / Doctor of Philosophy

No description available.

No description available.