주요 내용으로 건너 뛰기
/ 자료 / 블로그 / Drug Discovery Platform Unlocks Scientific Data Access

Drug Discovery Platform Unlocks Scientific Data Access

The D360 desktop application supports drug discovery workflows, data access, and early development research data analysis. D360 is also an organization-wide solution that provides powerful server features. D360 server’s data virtualization and the client application’s user interface meet scientists’ needs by

  1. Enabling workflow automation,
  2. Improving data analysis consistency, and
  3. Reducing IT infrastructure costs.

What about drug discovery scientists requesting data access but who don’t need a graphical user interface? How do they fit into the D360 community?

The D360 REST API and programmatic access

D360’s Representational State Transfer (REST) API is an appealing entry point for data scientists. This serves as a communication layer between the desktop application and the server. The REST API is based on the REST architectural style and HTTP(S).

It allows developers to build scripts and applications based on the D360 server and the data it can access. The REST API exposes most D360 operations and provides additional endpoints for external workflows.

D360 offers easy and controlled access, empowering all interested users to perform various tasks programmatically. Most modern programming languages can support the REST API framework. D360’s Python API wrapper package provides a convenient and familiar interface in the data scientists’ language of choice.

D360 unlocks access to configured data sources which scientists can use via the REST API.
Figure 1 – D360 server unlocks data access to all configured data sources. Data scientists can use them programmatically via the REST API.

Advantages of using D360 for drug candidates’ data access

D360 is extensible and configurable. The software allows organizations to add multiple data sources such as relational databases, cloud database systems, and web services. D360’s data virtualization layer generates a data catalog that provides an overview of the available data in an understandable format. Organizations can customize this catalog to show relevant items and ensure data security.

Users can utilize D360’s data catalog and self-service queries to access data no matter where it resides. Built-in query optimization and performance tuning allow easy implementation of external scripts and programs for data retrieval.

The 80/20 rule of drug design data science

Data scientists spend significant time on data retrieval and processing. This prep work can consume up to 80% of the data scientists’ time, leaving only 20% for generating insights.

D360 streamlines data processing in several ways. It applies standardized data cleaning practices, unit handling, data type recognition, and aggregation control specifically designed for drug discovery research data. Because D360 simplifies data preparation, scientists can focus on high-value tasks such as modeling and machine learning.

Insights, high throughput data visualization, and collaboration in D360

Another key advantage of using D360 as the data engine is its built-in collaboration capabilities. Since users build queries interactively in the desktop application, experts can collect and control the data. Furthermore, anyone accessing the query can explore the dataset with D360’s visualization and analytical capabilities.

Data scientists can create figures through programming, but many rely on a graphical user interface to generate them. Data scientists also can push predictions and other results for others to evaluate via the REST API. Democratized access to data and insights across different technical skill levels is vital to facilitate communication and collaboration during drug development.

The future of D360 is AI-driven

D360 has always focused on bringing data together from multiple sources including databases, web services, and cloud storage systems. By utilizing the REST API, in-house developers can build applications on top of the D360 server. For example, this technology can be used to implement specialized visualization or customized reporting tools for the organization.

D360 is working with Certara.AI to bring state-of-the-art large-language models, unstructured data searches, and de novo compound generation to users. This adaptable pharmaceutical research solution helps translate computational insights into informed business decisions.

To learn more about D360’s self-service query building and get examples of how it can transfer data between external applications, watch this webinar.

About the author

Dr. Essi Koskela
By: Dr. Essi Koskela

Dr. Essi Koskela is an Enterprise Informatics Consultant at Certara. After obtaining her Master’s degree in Biochemistry and PhD in Biotechnology, she has spent several years working with scientific software and consultancy services for the life science industry. She focuses on delivering smart data solutions for drug discovery and development.

Powered by Translations.com GlobalLink Web Software