Public Funding for Precisionlife Ltd

"The ORAC project will develop a new head-up display for scientists to navigate complex information spaces and improve the efficiency and accuracy with which they access scientific knowledge. It will provide scientists with a simple and comprehensive tool to correlate, annotate and navigate the widest range of data overcoming existing barriers between data sources. This will be delivered seamlessly within their current browser via a simple plug-in that identifies concepts on any HTML/PDF page and annotates them with semantically and contextually accurate information. The ORAC plug-in will automatically identify key scientific concepts such as 'disease', 'gene', 'chemical', 'bioassay', 'person' and link these to other resources for further study, hypothesis generation and verification. These will be derived from a wide range of data sources including definitions, attributes, supplemental datasets, and documented relationships to other types of concepts. This will enable much faster discovery and interpretation of information, and navigation between concepts whose descriptions may be held in different data sources. ORAC will take advantage of a comprehensive semantically normalized knowledge graph of all publicly available biomedical literature, including PubMed, Arxiv, USPTO, ChEMBL, OMIM, ClinicalTrials.gov, DBPedia etc. This knowledge graph will be automatically built and maintained by extensions to RowAnalytics' highly innovative spot.my deep semantic learning engine. RowAnalytics' spot.my deep learning engine provides major advantages over existing keyword, AI or NLP systems, as it learns the semantic meanings of text as documents are indexed, through detailed analysis of the patterns of co-occurrence of all concepts with all other concepts, and their distribution across all documents in a corpus. This makes it incredibly simple, efficient and scalable in use, and highly adaptable to a broad range of emergent new scientific terms. In the ORAC project, this will be extended to enable automated ontology construction and curation without expensive, time-consuming and inaccurate manual intervention or continually retraining of neural networks. The ORAC system will also provide personalized knowledge channels which users can set up around any topic to summarise and/or alert them to new information relevant to their subjects of interest. These unprecedented 'deep personalization' capabilities deliver smart AI searching in a fully private and secure manner with no sharing of personal data. They enable exploration of the scientific literature by biomedical researchers and clinical practitioners, for example to contextualize advice for a specific individual based on their combination of genomic, clinical and phenotypic attributes."

To develop a novel IoT edge analytics platform for recording human or animal activity using complex analytics to interpret signal data locally and provide real-time contextualised responses; providing secure and personalised monitoring of e.g. vulnerable people, using low-cost IoT devices.

Title: PROOF-IT Abstract: The PROOF-IT Phase 1 project will apply deep semantic learning and indexing technology on the provided Innovate UK Funded Projects dataset and existing publicly available Innovate UK guidance and advice documents. It will use the same underlying semantic learning engine to provide proofs-of-concept for several use cases, including: 1. automated tools for more efficient and accurate assessor allocation, 2. rapid, automated detection of undeclared resubmissions, 3. a partnering portal, to rapidly identify potential collaborative R&D project partners, and 4. more accurate answering of users’ natural language questions in FAQ/support/enquiry forum settings There will be 3 proof-of-concept interfaces demonstrated in Phase 1 - an integrated grant application centric system combining use cases 1 & 2, and two search interfaces for use cases 3 and 4. The PROOF-IT system will be built on RowAnalytics’ existing spot.my deep semantic learning system, developed in part with Innovate UK support, which has been used to efficiently semantically analyse, index and search large text collections such as scientific, patent & legal literature, e-retail listings to identify brand protection violations, image collections for predicting the effectiveness of digital advertising and even complex representations such as chemical structure data. The spot.my deep semantic learning engine requires no training and so is much more efficient, flexible, cheaper and easier to operate & use than other Machine Learning and AI approaches. While learning the deep semantic relationships between concepts, crucially it does not require re-training or curation of new controlled vocabularies when new or previously unencountered technical domains and keywords are required to be matched. It can semantically index huge document sets in minutes using standard GPU computing, and the system support tens of millions of keywords and billions of documents. Document sets can be continually updated using incremental indexing to make information available within minutes of it becoming available. Phase 2 of the project will build on these applications, scaling out the document sets to covered to the full text of all applications received and automating full semantic equivalence checks for undeclared resubmissions (even when wording has been changed to obfuscate similarities). An assessor allocation tool will also be built based on semantic matching of the set of available assessors’ resumes to the received applications for a call and knowledge of other attributes for the assessors that Innovate UK thinks relevant, e.g. number/type of existing projects, conflicts of interest, location etc. The user facing information search tools could also be extended to include context sensitive and semantically powered chatbots for dealing with routine user enquiries, support requests and general Q&A. The semantic engine would also provide a substrate for detailed analysis and correlation of projects’ thematic and/or application features with outcome metrics such as project forecasting accuracy and monitoring reporting scores.

ASSIST will build a new smart sensor platform that provides a step change in on-sensor analytical capability. The complex analytics needed to integrate multi-dimensional signal data from multiple sensors, accurately interpret them and provide a useful contextualized response will all be embedded in the sensor platform itself. This enables more personalized and informative responses to be delivered directly to the end user. The platform will drastically lower power usage and data transmission rates, enabling smaller, less intrusive devices, and increasing the user’s security. ASSIST will integrate new sensor, security and analytics technologies to provide a disruptive IoT platform that has applications across a range of internationally important markets such as healthcare, food and security. The ASSIST project will demonstrate the use of the platform help aging consumers live safely at home in a way that preserves their health and independence and delivers peace of mind for their family members and caregivers. It will be significantly cheaper and simpler than existing systems, will be non-intrusive and secure in use and will provide better outcomes for its users.

The COMBIO project will develop a precision medicine delivery platform to help clinicians evaluate a patient’s individual disease risks and their optimal therapeutic & dietary regimens, in real-time at the point of care. The project is underpinned by RowAnalytics’ groundbreaking combinatorial GWAS system and personalized dietary advisor tool. Large (>15,000 patient) multi-factor (genomic, phenotypic, clinical and epigenetic) datasets from MND/ALS and breast cancer will be used to exemplify the potential of COMBIO for delivering precision medicine advice at the point of care in routine clinical practice. RowAnalytics’ extensive combinatorial association studies on genomic, phenotypic & clinical data have identified well-differentiated and reproducible sub-populations of patients sharing common clusters of up to 17 factors in combination, which provide for much more personalised recommendations than existing gene tests, e.g. BRCA1/2. COMBIO will develop clinical decision support tools applying these biomarkers clusters to enable stratification of patients, disease risk scoring and personalisation of prescription & dietary advice on mobile devices.

The Lumen Content, Interaction, Attention and Rewards (LUCIAR) system is a digital interaction platform that measures the attention that consumers give to digital content, puts a price on that attention and communicates the value of that attention and interactions to them. LUCIAR will enable and incentivise content providers, retailers, brands and consumers to share the value created from quantification of consumers’ interactions with and responses to digital content. LUCIAR uses eye tracking hardware and sophisticated modelling of attention to evaluate consumers’ responses to specific pieces of digital content such as ads, webpages and social media. These are converted into predicted buying behaviour using Lumen’s proven econometric response models. The measurement of this underlying attention and response can be used by retailers and brands to directly assess the likely impact of content designs in specific demographics. At the same time motivated consumers can control the sharing of their interaction data in return for a series of tangible, personalised rewards.

This project will build a demonstrator to support crop yield improvement by integrating and exploiting plant phenotyping and ‘omics data. This is a strategically important area of unmet need both in the UK and globally, where optimising the combination of inputs, treatments, doses & timings to apply to specific crop strains in a specific growing environment is a massively complex challenge. We will build a semantic knowledge graph of crop strains, traits (both genomic variant and from integrated non-destructive phenotyping technologies), yields, soil conditions and wider growing environment including remote sensing data. We will compile this knowledge model and deploy it as a demo tablet app with a simple decision support UI to allow non-IT specialist growers to identify specific combinations of treatments that enable them to optimise crop yields and minimise wastage of inputs.