Deriving mammalian retina from stem cells has had a large impact on the study of the biology of vision and is called organoid. Compared to in vivo retina, retinal organoids are far less functionally sophisticated in terms of their synapses, connectivity, discrimination between different light stimuli and their electrical action potentials. This project will overcome this functional constraint of retinal organoids by studying electrophysiological events-derived functional maturation of mouse retina during retinal development and then stimulating those events with the help of mathematical models in order to induce the same functionality in mouse and human retinal organoids. NeuFRO will achieve a resonance in the field by generating retinal organoids with the neuronal connectivity and the natural diversity of functions using interdisciplinary fields including electrophysiology, developmental biology, and computationally-derived electrical stimulation. Initially, I will create a holistic roadmap of the electrical features of immature mouse retina during development that shows self-organization through electrophysiology. With milli- to nanometer imaging precision, electrical activities derived the circuit formation will be spatiotemporally documented. Then I will decode this space-time code of intrinsic electrical patterns and neuronal connectivity using an ambitious strategy incorporating Hodgkin-Huxley and linear-nonlinear models. Next, such electrical response models will be applied to immature retinal organoids (mouse and human) by an innovative ‘sandwich’ electrophysiology technique during the development in vitro. With this approach, I will induce naturalistic electrical features in the retinal organoid, allowing the functional neurons to wire and fire appropriately into retinal organoids, particularly visual circuits. This ground-breaking approach will advance techniques for generating functional human retina.

Cancer is the second leading cause of mortality in EU member states with ~90% of all cancer deaths caused by metastatic spread. Despite its significance, measuring metastatic potential as well as potential indicators of therapy efficacy remain unmet clinical challenges. Recently, it has been demonstrated in vitro, that aggressive metastatic cells pull on their surroundings suggesting that metastatic potential could be gauged by measuring the forces exert by tumours. Furthermore, many solid tumours show a significantly increased interstitial fluid pressure (IFP) which prevents the efficient uptake of therapeutic agents. As a result, a reduction in IFP is recognized as a hallmark of therapeutic efficacy. Currently, there is no non-invasive modality that can directly image these forces in vivo. Our objective is the non-invasive measurement of both IFP within tumours as well as the forces they exert on their surrounding environment. This will be used to predict a tumour’s metastatic potential and importantly, changes in these forces will be used to predict the therapeutic efficacy of drug therapy. To attain this goal, the biomechanical properties of the tumour and its neighbouring tissue will be measured via MR-elastography at various measured deformation states. Resultant images will be used to reconstruct images of the internal and external forces acting on the tumour. We call this novel imaging modality Magnetic Resonance Force (MRF) imaging . We will calibrate MRF via cell cultures and pre-clinical models, and then test the method in breast, liver, and brain cancer patients. Thereby, we will investigate whether MRF data can predict metastatic spread and measure IFP in patients. We will also investigate the potential to non-invasively modulate the force environment of cancer cells via externally applied shear forces with the aim of impacting cell motility and proliferation. This can provide novel mechanism for anticancer therapeutic agents via mechanotransduction.

The rapid growth of web data creates demand for software engineering methods which can build and maintain applications that extract, process and publish web data. However, converting Big Data sources into high-quality, structured knowledge for use in business processes is usually considered data engineering. ALIGNED will develop models, methods and tools for engineering information systems based on co-evolving software and web data. Tools for model-driven software evolution based on Linked Data sources, runtime data quality analytics, human data curation and process integration will aid more efficient governance, increased agility and higher productivity. The opportunities for web data have recently led to intense research and innovation, but most efforts are siloed in the software or data spaces. Past integration of data and software engineering used formal ontologies rather than Linked Data. ALIGNED will provide lightweight methods for European data and software engineering industries to exploit the new opportunities in web data. Information companies like Wolters Kluwer need better ways to extract web data and build applications on top of it. Public bodies like the UK National Health Service (NHS) need evolving systems for data collection and re-use. Web data publishers like DBpedia need new methods to improve data quality. Scientific publications, industry workshops, training programs, open source tools and engaging the OMG, W3C and ISO standards bodies will transfer ALIGNED outputs. ALIGNED combines world class researchers in model driven software engineering (Oxford are transforming NHS systems), Linked Data quality (Leipzig and Trinity College have published foundational papers) and web systems (Leipzig are co-creators of DBpedia) with innovative enterprises (Wolters Kluwer have Linked Data in production systems, Semantic Web Company lead the world in enterprise Linked Data) and pioneering expert-curated data publishers (Oxford Anthropology and Posnan).

The human brain is flexible. Neural networks adapt to cognitive demands by flexibly recruiting different regions and connections. Flexible network adaptation enables cognitive functions such as semantic cognition: the ability to use, manipulate, and generalize knowledge. When key nodes suffer damage, networks can adapt to recover function. Yet, brain lesions often severely impair semantic cognition. How the semantic network adapts to lesions is poorly understood. My hypothesis is that disruption of the semantic network can be compensated for by recruitment of domain-general networks. This notion is based on findings that disruption of semantic nodes inhibits semantic activity but increases activity in domain-general nodes. Yet, the behavioral relevance of domain-general recruitment is unclear. Compensation means that behavior can be preserved as other nodes work harder. Can domain-general networks effectively compensate for disruption of specialized nodes? Is this a common principle of flexible adaptation in the healthy young, aging, and lesioned brain? Unprecedented inhibitory and facilitatory neurostimulation will be used to unbalance and rebalance network adaptation in semantic cognition. Importantly, a novel network stimulation approach will target multiple nodes simultaneously. I ask three questions. (i) Can domain-general networks compensate for semantic network disruption? (ii) Is domain-general recruitment in the aging brain adaptive? (iii) Do domain-general networks drive flexible adaptation to lesions? Perturbing young brains will elucidate the relevance of network adaptation. Perturbing aging brains will probe compensatory reorganization. Facilitating lesioned brains will reshape flexible adaptation. Benefitting from my strong neurostimulation experience, we will elucidate the way the brain compensates for disruption. The potential impact of the project on current conceptions of brain plasticity, and for rehabilitative medicine in particular, is immense.