NETWORK TOPOLOGY IN HUMAN PROTEIN INTERACTION DATA PREDICTS FUNCTIONAL ASSOCIATION

High-throughput assays, such as yeast two-hybrid system, have generated a huge amount of protein-protein interaction (PPI) data in the past decade. This tremendously increases the need for developing reliable methods to systematically and automatically suggest protein functions and relationships between them. With the available PPI data, it is now possible to study the functions and relationships in the context of a large-scale network. To data, several network-based schemes have been provided to effectively annotate protein functions on a large scale. However, due to those inherent noises in high-throughput data generation, new methods and algorithms should be developed to increase the reliability of functional annotations. Previous work in a yeast PPI network (Samanta and Liang, 2003) has shown that the local connection topology, particularly for two proteins sharing an unusually large number of neighbors, can predict functional associations between proteins, and hence suggest their functions. One advantage of the work is that their algorithm is not sensitive to noises (false positives) in high-throughput PPI data. In this study, we improved their prediction scheme by developing a new algorithm and new methods which we applied on a human PPI network to make a genome-wide functional inference. We used the new algorithm to measure and reduce the influence of hub proteins on detecting functionally associated proteins. We used the annotations of the Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) as independent and unbiased benchmarks to evaluate our algorithms and methods within the human PPI network. We showed that, compared with the previous work from Samanta and Liang, our algorithm and methods developed in this study improved the overall quality of functional inferences for human proteins. By applying the algorithms to the human PPI network, we obtained 4,233 significant functional associations among 1,754 proteins. Further comparisons of their KEGG and GO annotations allowed us to assign 466 KEGG pathway annotations to 274 proteins and 123 GO annotations to 114 proteins with estimated false discovery rates of <21% for KEGG and <30% for GO. We clustered 1,729 proteins by their functional associations and made pathway analysis to identify several subclusters that are highly enriched in certain signaling pathways. Particularly, we performed a detailed analysis on a subcluster enriched in the transforming growth factor β signaling pathway (P<10-50) which is important in cell proliferation and tumorigenesis. Analysis of another four subclusters also suggested potential new players in six signaling pathways worthy of further experimental investigations. Our study gives clear insight into the common neighbor-based prediction scheme and provides a reliable method for large-scale functional annotations in this post-genomic era.

ΔNp63 REGULATES A COMPLEX NETWORK OF TARGET GENES IN LIMB AND EPIDERMAL DEVELOPMENT

The skin is composed of two major compartments, the dermis and epidermis. The epidermis forms a barrier to protect the body. The stratified epithelium has self-renewing capacity throughout life, and continuous turnover is mediated by stem cells in the basal layer. p63 is structurally and functionally related to p53. In spite of their structural similarities, p63 is critical for the development and maintenance of stratified epithelial tissues, unlike p53. p63 is highly expressed in the epidermis and previously has been shown to play a critical role in the development and maintenance of the epidermis. The study of p63 has been complicated due to the existence of multiple isoforms: those with a transactivation domain (TAp63) and those lacking this domain (ΔNp63). Mice lacking p63 cannot form skin, have craniofacial and skeletal defects and die within hours after birth. These defects are due to the ability of p63 to regulate multiple processes in skin development including epithelial stem cell proliferation, differentiation, and adherence programs. To determine the roles of these isoforms in skin development and maintenance, isoform specific p63 conditional knock out mice were generated by our lab. TAp63-/- mice age prematurely, develop blisters, and display wound-healing defects that result from hyperproliferation of dermal stem cells. That results in premature depletion of these cells, which are necessary for wound repair, that indicates TAp63 plays a role in dermal/epidermal maintenance. To study the role of ΔNp63, I generated a ΔNp63-/- mouse and analyzed the skin by performing immunofluorescence for markers of epithelial differentiation. The ΔNp63-/- mice developed a thin, disorganized epithelium but differentiation markers were expressed. Interestingly, the epidermis from ΔNp63-/- mice co-expressed K14 and K10 in the same cell suggesting defects in epidermal differentiation and stratification. This phenotype is reminiscent of the DGCR8fl/fl;K14Cre and Dicerfl/fl;K14Cre mice skin. Importantly, DGCR8-/- embryonic stem cells (ESCs) display a hyperproliferation defect by failure to silence pluripotency genes. Furthermore, I have observed that epidermal cells lacking ΔNp63 display a phenotype reminiscent of embryonic stem cells instead of keratinocytes. Thus, I hypothesize that genes involved in maintaining pluripotency, like Oct4, may be upregulated in the absence of ΔNp63. To test this, q-RT PCR was performed for Oct4 mRNA with wild type and ΔNp63-/- 18.5dpc embryo skin. I found that the level of Oct4 was dramatically increased in the absence of ΔNp63-/-. Based on these results, I hypothesized that ΔNp63 induces differentiation by silencing pluripotency regulators, Oct4, Sox2 and Nanog directly through the regulation of DGCR8. I found that DGCR8 restoration resulted in repression of Oct4, Sox2 and Nanog in ΔNp63-/- epidermal cells and rescue differentiation defects. Loss of ΔNp63 resulted in pluripotency that caused defect in proper differentiation and stem cell like phenotype. This led me to culture the ΔNp63-/- epidermal cells in neuronal cell culture media in order to address whether restoration of DGCR8 can transform epidermal cells to neuronal cells. I found that DGCR8 restoration resulted in a change in cell fate. I also found that miR470 and miR145 play a role in the induction of pluripotency by repressing Oct4, Sox2 and Nanog. This indicates that ΔNp63 induces terminal differentiation through the regulation of DGCR8.

ADHERENS JUNCTIONS AND THE ACTOMYOSIN NETWORK REGULATE ORGAN GROWTH BY MODULATING HIPPO PATHWAY ACTIVITY IN DROSOPHILA

Adherens junctions (AJs) and basolateral modules are important for the establishment and maintenance of apico-basal polarity. Loss of AJs and basolateral module members lead to tumor formation, as well as poor prognosis for metastasis. Recently, in mammalian studies it has been shown that loss of either AJ or basolateral module members deregulate Yorkie activity, the downstream transcriptional effector of the Hippo pathway. Importantly, it is unclear if AJ and basolateral components act through the same or parallel mechanisms to regulate Yorkie activity. Here, we dissect how loss of AJ and basolateral components affects Hippo signaling in Drosophila. Surprisingly, while scrib knock-down tissue displays increased reporter activity autonomously, α-cat knock-down tissue shows a cell autonomous decrease and a cell non-autonomous increase of Hippo reporter activity. We provided several lines of evidence to show the differential regulation in polarity protein localizations and oncogenic cooperative overgrowth by AJs and basolateral complexes. Finally, we show that Hippo pathway activity is induced in α-cat and scrib double knocked-down tissue. Taken together, our results provide evidence to show that basolateral modules and AJs act in parallel to modulate Hippo pathway activity. Non-muscle myosin II is an actomyosin component that interacts with the actin. Non-muscle myosin II also interacts with lgl, though the function of this interaction is not clear. Our lab demonstrated that modulating F-actin regulates Hippo pathway activity, and lgl also has been described as a Hippo pathway regulator. Therefore we suspect that myosin II is also involved in Hippo pathway regulation. We first characterized non-muscle Myosin II as a novel tumor suppressor gene by affecting Hippo pathway activity. Upstream regulators of Myosin II, members in the Rho signaling pathway, also displayed similar phenotypes as the Myosin II knock-down tissues. Apoptosis is also induced in myosin II knock-down tissues, however, blocking cell death does not affect myosin II knock-down induced Hippo activation. Our data suggested hyperactivating myosin II induced F-actin accumulation so therefore induces Hippo target activation. Unexpectedly, we also observed that reducing F-actin activity induced Hippo target activation in vivo. These controversial data indicated that actomyosin may regulate the Hippo pathway through multiple mechanisms.

Network-based analysis of genome-wide scans of natural selection

The genomic era brought by recent advances in the next-generation sequencing technology makes the genome-wide scans of natural selection a reality. Currently, almost all the statistical tests and analytical methods for identifying genes under selection was performed on the individual gene basis. Although these methods have the power of identifying gene subject to strong selection, they have limited power in discovering genes targeted by moderate or weak selection forces, which are crucial for understanding the molecular mechanisms of complex phenotypes and diseases. Recent availability and rapid completeness of many gene network and protein-protein interaction databases accompanying the genomic era open the avenues of exploring the possibility of enhancing the power of discovering genes under natural selection. The aim of the thesis is to explore and develop normal mixture model based methods for leveraging gene network information to enhance the power of natural selection target gene discovery. The results show that the developed statistical method, which combines the posterior log odds of the standard normal mixture model and the Guilt-By-Association score of the gene network in a naïve Bayes framework, has the power to discover moderate/weak selection gene which bridges the genes under strong selection and it helps our understanding the biology under complex diseases and related natural selection phenotypes.^