Read about FAIR data principles - a relatively new concept for data discoverability and management that has quickly gained traction among the scientific data community…
In our previous post dedicated to Ontotext webinars, we talked about how our partner technology ecosystem can deliver powerful knowledge graph based end-to-end solutions. To put it in a nutshell, we believe that only such an ecosystem can stack enough complementary capabilities to deliver efficient end-to-end solutions. We also believe that often knowledge graph driven approaches not only address the business need, but redefine the problem, opening up new opportunities as enterprises change their knowledge management processes.
Whether operating in Pharma & Life Sciences, Engineering & Manufacturing, Financial Services or another industry vertical, large enterprises have identified knowledge graphs as the foundation for making data FAIR and for unlocking the value of their data assets. However, for Pharma & Life Sciences, knowledge graphs built on FAIR data are especially important to stay ahead of the competition. They drive digital transformation initiatives and allow companies to address topics such as precision medicine, full DNA sequencing, digital therapeutics, etc.
With this framework in mind, we want to present a couple of webinars in the domain of Pharma & Life Sciences, which we have done with another one of our partners – metaphacts. We are happy to collaborate with metaphacts as together we are greater than the sum of our parts and thanks to our joint capabilities, we can address many use cases in the drug lifecycle. From the diagram below, you can get an idea of the potential of our joint technology, but keep in mind that it is not limited to the use cases on it nor to the specified domain.
In the Research and Discovery phase, we serve global pharmaceutical companies in use cases like therapeutic target discovery and we help innovative biotech companies with their drug repurposing. In Preclinical Development, we build knowledge graphs for hypothesis testing or sample management. We have a lot of use cases in Manufacturing and especially such related to regulatory compliance. We also have a lot of experience in Clinical Trials with use cases in scientific communication, regulator intelligence or clinical trial recruiting. Finally, in the Marketing and Distribution phase, we have worked on identifying key opinion leaders as well as medical inquiry analytics and product labeling and updates.
Currently, Ontotext and metaphacts approach is being used productively in many customer implementations. It has helped our customers successfully adopt knowledge graphs as a strategic tool to drive digital transformation across a variety of use cases and has led to significant boosts in productivity, cost reductions and efficiency optimization.
Read on and watch the webinars to get to the details!
Our first webinar Generate Value from Your Knowledge Graph in Days is presented by Todor Primov, Life Sciences and Healthcare Solutions at Ontotext, and Sebastian Schmidt, CEO at metaphacts. After a brief introduction of the two companies, the webinar focuses on how, if you have already built a knowledge graph with GraphDB, you can use the metaphactory knowledge graph platform on top of GraphDB to accelerate your knowledge graph journey. Despite the popularity of knowledge graphs, the road to their implementation has often been long and complex, and success has relied on the involvement of seasoned knowledge graph experts.
Todor and Sebastian use an example from clinical trial scoping to demonstrate how our two products work together and how you can use the platform’s components to quickly gain value from your knowledge graph. You learn how a pharmaceutical company has leveraged metaphactory on top of GraphDB to build a user-oriented dashboard that brings together data on hundreds of thousands of clinical studies, investigators, sites, medical conditions, and their symptoms, drugs and common adverse effects, etc. This dashboard makes it easy for clinical study managers to identify the optimal site for their study by looking at available institutions and their locations, the subjects available in each respective area, the indications commonly studied at these institutions, etc.
You can also read our joint blog post based on this webinar: GraphDB & metaphactory Part I: Generating Value from Your Knowledge Graph in Days & GraphDB and metaphactory Part II: An RDF Database and A Knowledge Graph Platform in Action.
Interesting attendee question: Do you support building clinical content of harmonized data before it can be used to visualize as a knowledge graph?Ontotext & metaphacts answer: Yes, this is exactly the approach we have followed in building the knowledge graph for the demo. Harmonization happens as part of the initial transformation and loading as we apply different techniques for semantic instance mappings (based on referential identifiers across data sets) and NLP for normalization of unstructured data (both listerals and large texts).
Our second webinar Accelerated R&D with a FAIR Data Fabric in Life Sciences features Ontotext’s Ilian Uzunov (Sales Director Life Sciences and Healthcare) and Todor Primov (Life Sciences and Healthcare Solutions) and metaphacts’s CEO Sebastian Schmidt. They talk about the benefits of knowledge graphs for driving digital transformation in Life Sciences & Pharma and share important lessons learned from many customer implementations.
You see examples from a pre-clinical knowledge graph capturing OMICs data (genomics, proteomics, transcriptomics) together with information about pathways, tissues, drugs and diseases. The knowledge graph used in the webinar is built entirely of publicly available datasets, including ChEMBL, Ensembl, UniProt, GWAS Catalog and Reactome. These datasets are semantically mapped to each other and against the common data model.
You also learn how to add your own data (e.g., in the form of expression data), how to get deeper insights (e.g., how to identify targets for new drugs through protein interaction partners or along the pathway) and how to answer very specific requests.
Interesting attendee question: How does GraphDB with metaphactory differentiate from other solutions for Life Science R&D?
Ontotext & metaphacts answer: The approach of GraphDB and metaphactory is to start out from a model of your domain, which is almost always based on a public ontology and then extend it or adjust it to fit the specific information needs you want to address. This ontology drives the data integration (structured, unstructured and ready-to-use data sources) as well as the creation of user interfaces that are additionally tailored to the required user experience for your end users. In our understanding, an approach that builds on open standards (W3C) is unique on the market.
Interesting attendee question: What is the method of enriching your knowledge graph?
Ontotext & metaphacts answer: We use the core of the knowledge graph (main classes and properties) to populate our NER pipelines. We use these NLP pipelines to semantically annotate any unstructured content (both documents and also textual fields from certain datasets) in order to identify the objects from the knowledge graph and the relations between them. We use the background knowledge in the knowledge graph to disambiguate the annotations in the specific context and to further provide additional information that will be helpful to interpret the information in the document. All of the extracted information from the unstructured content is transformed into RDF based on the annotation schema and semantically fused with the core knowledge graph. Each fact in the knowledge graph that was extracted from text can be traced back to its provenance source (document > section> sentence > off set).
That covers the two webinars that we wanted to present to you today.
To sum it up, our extensive partner technology ecosystem effectively covers the requirements of end-to-end enterprise solutions (dive in Atanas Kiryakov’s webinar: Knowledge Graph Maps: 20+ Application and 30+ Capabilities for a comprehensive overview of the different applications of knowledge graphs and the capabilities that enable them). The blend of our technologies provides numerous advantages and allows enterprises to move towards data fabrics built on FAIR data that offer a modular approach to data integration and knowledge consumption.
We hope you’ve heard enough to want to dive straight into one of them or why not all three?
Watch out for our next post in this series At Center Stage or visit our Webinars directly to check out what is upcoming!