Biovista's pharma services, software solutions and drug repositioning work are all supported by a broad technology platform that is both extremely powerful and flexible. This proprietary platform (called Discovery Algorithmics^TM) uses four major A.I. components:

• Natural Language Processing (to extract information fro unstructured data sources)
• Ontologies (to organise information in semantically useful categories)
• Visualization (to help understand complex data sets)
• Data and Text Mining (to identify non-obvious information from very large data sources)

The platform is deployed over Biovista's growing internal database of scientific articles from major journals, patents, and news items. With over 18 million records and 4 billion rows it is currently the 8 largest MySQL database in the world. Discovery Algorithmics^TM can extract actionable information from Biovista's databases in runs that take from seconds to a few minutes. Some of the ontologies used by the system include:

• All FDA-approved drugs
• Medical tests and procedures
• Laboratory experiment types and applications
• Reagents
• All reported human genes
• Signal transduction pathways
• All reported genetic and non-genetic human diseases

Biovista's technology components are combined into specific workflows, to provide turn-key solutions that maximize the value of research and clinical development.

We have combined the best of opinion-based medicine with advanced knowledge automation. Using the mode of action of a drug, an Adverse Event or a disease, characterized by all possible data from molecular epidemiology, expression profiling and context-based pathophysiology. Biovista links any drug via its mode of action to previously unknown alternative indications or AEs, and then generates the relevant clinical development plan to execute. Biovista’s technology and development workflow are based on three components working together to deliver an action plan:

1. Meta-review and Gap Analysis of the disease molecular epidemiology. Biovista begins with this, as it allows for knowledge gaps to be identified. For example, in pancreatic cancer, there are over 24 molecular markers that are differentially expressed in patients and that need to be inhibited by drugs. No drug inhibits all 24, and therefore a combination needs to be identified base on this variable expression profiling.
2. Profiling and Screening. Using Biovista’s proprietary 4 billion row association database accessed through advanced AI-based analytics, Biovista uses the molecular epidemiology of a drug to link it to potential new indications or side effects. For example, in pancreatic cancer Biovista has identified a number of drugs that have overlap coverage for the pancreatic cancer markers and are not currently used for the purpose.
3. Clinical Development Plan or Risk Mitigation Plan (RMP). Combining the gap-analysis and predictive profiling, we produce the Clinical Development Plan and RMP on demand. These would be used as implementation steps for a drug in preclinical or clinical phases. For example, Biovista has generated a new drug combination plan for pancreatic cancer that incorporates new drugs in combination with existing therapies, to enhance the coverage of the pancreatic cancer markers as they are expressed in patients.

It takes Biovista 30-45 days to de-risk and reposition a drug, including full biological plausibility and the outline of the clinical development plan or RMP. A full Clinical Development Plan development may take an additional 30-45 days. This rapid timeline is a major advantage for Biovista and our partners.

To see a list of selected publications by Biovista personnel, click here.

To find out more about the technologies used in Discovery Algorithmics^TM platform, select from the menu on the left or contact us directly at info@biovista.com.

To find out more about the use of these technologies in company projects, you may visit the Research page of this site.