The proposed research is centred around statistical analysis of dynamic multiplex graphs (DMPGs). Mathematically, a graph, also known as network, can be interpreted as a collection of nodes, with edges occurring between them. Network data are collected in many domains, such as healthcare, biology, and cyber-security, and they are becoming increasingly rich, continuously generating new research questions. In particular, dynamic multiplex networks are emerging as increasingly common data structures observed in real-world applications. In DMPGs, edges could have different types, and evolve in time. For example, in an enterprise computer network, nodes could be represented by hosts, and edges correspond to connections between them, occurring dynamically over time on different ports. Because of the complexity of such objects, research has only scratched the surface with statistical modelling for DMPGs. Therefore, the development of novel statistical methodology is required, and this research intends to bridge this gap, developing realistic statistical models for DMPGs. The aim of this research proposal is to develop principled and scalable statistical models which represent the full multi-layered complexity of dynamic multiplex graphs. This goal will be achieved by exploiting an array of statistical techniques, spanning from spectral methods to topic modelling. In particular, this research proposal focusses on techniques for discovering low-dimensional substructure in networks, known as embedding methods. Such techniques have the added benefit of aiding subsequent inference tasks, such as clustering of nodes with similar behaviour. The statistical properties of novel embedding methods proposed for DMPGs will be carefully assessed, and the proposed methods will be utilised to improve and extend existing models for clustering, link prediction, and anomaly detection. In addition, the proposed models will have the flexibility to encompass additional information on nodes and edges, available in the form of covariates. In particular, this research proposal will focus on incorporating unstructured data, such as text, within the proposed modelling frameworks, combining aspects from network analysis and natural language processing.

The proposed research effort will support an international collaboration between the Chayen group at Imperial College, London (IC), specializing in protein crystallization and the Erlebacher group at Johns Hopkins University, Baltimore (JHU), which specializes in the synthesis and surface modification of nanoporous metals. By marrying these specializations, a fundamental study of heterogeneous crystal nucleation on nanoporous substrates will be efficiently pursued. The collaboration is structured with significant exchange of students and junior personnel; ultimately, the JHU team will develop in-house skills in solution-phase crystallization, and the IC group will learn materials synthesis techniques for fabricating nanoporous metals and nanoporous metal composites, leading to advances in crystallization generally. Fundamentally, our proposed research considers a new approach to thinking about the heterogeneous nucleation of macromolecular species. In traditional heterogeneous nucleation theory, nucleation rates are controlled by the relative interfacial energies between the solvent, substrate, and nucleus. Here, we will tailor the substrate to expose not just one uniform surface to the solution, but instead to expose a statistically relevant distribution of three-dimensionally oriented nucleation sites. Modifications of nucleation theory to explore this idea is an important part of the proposed work and is required for applying the technique generally.A detailed study of the nucleation rate of macromolecules over porous substrates will be performed using a new substrate material / nanoporous gold Nanoporous gold is a random, uniform, mesoporous material whose mean pore size can be varied from 3-50 nm using simple chemistry. Its surface is also easily chemically modified, and all these features together make it uniquely perfect for this kind of study.The results of this research will yield new understanding of the fundamentals of heterogeneous nucleation, with general applicability to protein and colloidal crystallization. Furthermore, the international collaboration will help to build international networks in nucleation theory and experiment, bridging the disciplines of thin film metallurgy and solution phase crystallization.

Myriad interactions between DNA and proteins that take place throughout the length of an organism's genome ultimately allow cells to read, repair, package, and copy DNA sequence. How cells properly orchestrate and control these critical DNA:protein interactions is a fundamental question in biology. A unifying theme across such diverse DNA:protein interactions is that they always require some form of local mechanical distortion of DNA like bending, twisting, or kinking. Therefore, DNA:protein interactions can potentially be modulated and controlled by the local mechanical properties of DNA such as its bendability. Structural studies, dynamic experiments, and computational works have suggested that the mechanical properties of the DNA polymer are not constant, but vary along its length depending on local sequence, via a "mechanical code". Over decades, this has given rise to the hypothesis that sequence may be able to significantly control the local mechanical properties of DNA, and via it, control the critical DNA:protein interactions that in turn allow sequence itself to be read, repaired, copied, and packaged. In other words, via the mechanical code, DNA sequence might be able to control its own regulation. If this hypothesis is true, because of its potential generality and likelihood of relevance in all examples of DNA:protein interactions in all organisms, it would represent a transformative step in our understanding of life and in our ability to control it. Towards this end, we recently developed high-throughput experimental methods to measure, for the first time, how the mechanical properties of DNA vary with sequence along large regions of the genomes of various organisms. Via other experiments, we showed that these sequence-encoded variations in DNA bendability regulate critical processes related to the reading, copying, and packaging of DNA. Genetic information in DNA sequence is further modified by chemical alterations to DNA such as methylation (addition of a methyl group mainly to the cytosine base of DNA). DNA methylation is of fundamental importance in altering which genes along DNA are expressed. While certain cellular factors have been found to recognise methylated DNA, how DNA methylation achieves so many broad downstream effects is not fully understood. Recently, it has been suggested that one of the ways in which DNA methylation could exercise control over DNA transactions is by modifying the local physical and mechanical properties of DNA. If true, DNA methylation might allow cells to dynamically alter the "mechanical code" itself, as a means of gaining a broad regulatory handle on many different DNA:protein interactions. A significant roadblock to exploring this hypothesis has been the lack of high-throughput methods to provide the basic characterization of how DNA methylation, at various points along an organism's vast genome, alter the local mechanical properties of DNA depending on local sequence context. Here we propose to extend the capabilities of our high-throughput experimental techniques to make it possible to characterize the mechanical consequences of DNA methylation in high-throughput throughout the genome. We will compare our findings with other genome wide data, and perform other high-throughput biochemical experiments on how DNA sequence and methylation affect protein:DNA interactions. We expect to develop a comprehensive understanding of how DNA methylation, via its impact on the local physical properties of DNA, impacts the local structure of chromatin and the expression of individual genes. As DNA methylation accompanies processes like embryonic development, cellular adaptation to environmental changes, and genetic diseases like cancers, this project lays the foundations for future efforts at understanding how such critically important processes in biology might, in part, achieve their effects by gaining a handle on the physical properties of DNA.

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

For the past decade, Sub-Saharan Africa has been growing, yet growth is not the same as structural transformation. China's development trajectory since 1980 provides an example of how a government focused on modernization can marshal foreign capital and technology to assist in the reduction of poverty and economic transformation in manufacturing and agriculture. In Africa, China is largely seen as a competitor for local firms, primarily through imports. This competition can be devastating in some countries and some sectors, driving local firms out of business. Yet on the other hand, growing Chinese investment in African manufacturing and contract farming can also offer opportunities for joint ventures with local firms, training, and diffusion of more productive technologies. If this were to follow Asian experience, Chinese firms could be catalysts for local firms to move into manufactured exports, although they might also be footloose investors, moving on with only fleeting impact on local knowledge. In agriculture, Chinese investment might also be enclave, with little connection to local farmers - the picture presented in fears of "land grabbing" - or it might follow the pattern laid out by foreign investors in China, with out-growers, demonstration farms, and technology and skills transfers. Our earlier research suggested that Chinese firms are thinking strategically about backward linkages. For example, at least five Chinese shoe manufacturers we interviewed in 2009 had moved their shoe-making assembly lines to Nigeria, while still importing uppers and soles from China. In 2012, one company was in discussions with their Chinese supplier about moving to Nigeria to produce soles locally from Nigerian rubber. Similarly, we have identified Chinese contract farming investments and commercial agriculture projects with demonstration farms, advisers, and input supplies in places like Mali, Zimbabwe, and Malawi. This project will enable a more refined picture of the actual scope and impact of Chinese investment and the potential and experience of technology transfer in commercial agriculture and agro-industry. We will combine multiple methods: database construction, scoping studies, cluster surveys, a national survey, and eight paired, comparative case studies, following an approach tested in our earlier research on Chinese agro-industrial and commercial agriculture engagement in Ethiopia (2011-2014), and Chinese commercial agricultural investment in Zambia and Zimbabwe (2013). The scoping studies will allow us to better map existing Chinese (and other) investment in agro-industry and commercial agriculture, while the cluster surveys will provide an overview of existing linkages and opportunities for technology transfer. A further level of depth will be obtained through adding a technology-transfer module to two national surveys of manufacturers. Finally, eight in-depth, paired case studies will complement the survey research by using process-tracing to compare specific experiences of agro-industrial FDI and technology transfer in China, with Chinese and a similar non-Chinese experience in Africa. For example, we will study the institutional framework and approach that allowed the Thai firm CP Group to become China's largest foreign investor in the Chinese poultry industry, with significant technology spinoffs, and compare this with the spinoffs and technology transfer from significant Chinese and South African investors in Zambia's poultry industry (Zhongken Farm and Astral Foods). The output of the research will be a far more robust basis for analysis of the current and future possibilities for technology transfer in China's African investment, and guidelines for governments and development partners to derive maximum benefit from these opportunities.