The field of data science is a fast evolving area. New technologies produce new and often larger data types which require either adaption of existing analytical methods, or the development of new methodologies. BSSI’s data scientists develop new strategies to utilize the vast real world data and evidence, reference population databases and proprietary client data to support drug target identification and development and other client objectives. Integrating different data types and sources and creating customized interfaces and visualization tools helps our clients to navigate the data and find meaningful relationships and results.
Examples of machine learning applications
- Prioritizing genes for drug target using a variety of evidence sources and annotations;
- Prediction of disease diagnosis or relapse after treatment using biomarker data for developing potential companion diagnostics or other diagnostic tests;
- Drug response and drug synergy prediction using DNA or RNA data and identifying patient subgroups with improved treatment effect;
- Support of clients in presenting machine learning and predictive modeling results to regulatory agencies.
“A picture is worth a thousand words”
This old adage is particularly true when considering the visual representation of data, especially complex and/or large data. An interactive image can help to see relationships, find trends or identify the “needle in the haystack”. At BSSI, we create flexible visualization tools that look at data from every angle, and use them as a means of communication between the analytical team and the scientist or end-user. Examples of expertise and experience include:
- Create production-level tables, figures and listings for submission/publication;
- Integrate data from multiple sources (e.g. clinical and molecular) with easy query tools at the front end;
- Develop exploratory visualization tools to allow for client’s interactive data exploration;
- Generate standalone reproducible reports.