With the increase of digitalization in Pharma R&D, increasingly large and complex system landscape have been established to address the manifold research use cases and processes. As a consequence, Pharma companies struggle with the exploded operative effort of their system landscapes limiting the potential for further innovation. In this talk, ongoing efforts and learnings around system landscape consolidation at Roche will be presented. On the example of molecule registration systems, concrete recommendations will be summarized how to approach the assessment and implementation of consolidating a specific business capability. At the same time, wrong assumptions and potential pitfalls around landscape harmonization will be shared with the audience.

Reimagining digital science in the age of AI.

With the growing number of public resources available, data preparation remains a challenging and time-consuming exercise for scientists. BioRels is an open-source fully tunable data preparation infrastructure processing 30 of the main drug discovery resources. We will present this infrastructure and its impact on Lilly Genetic Medicines drug design. We will also show how complex and heterogenous data can easily be shared between BioRels systems and what the future can look like if we expand it as a community.

Eisai’s Target Gene Notebook serves two primary functions. Firstly, it automates the collection of data from various informatics resources. Secondly, it offers a toolkit for subject-matter experts to compile and analyze this data, transforming it into actionable reports. These activities provide key genetic evidence for project target selection and safety evaluation.

We standardized and automated biospecimen and CAR construct management, ensuring FAIR data principles. Key achievements include defining data models, migrating thousands of biospecimens, creating templates, automating calculations, developing workflows, and implementing sequence identification, reducing manual annotation time by 60%. These efforts improved data quality, searchability, and collaboration, demonstrating the value of innovative informatics solutions in pharmaceutical research.

Today’s R&D laboratory generates vast amounts of diverse data acquired with heterogeneous hardware and software solutions supplied by various vendors. The heterogeneity in vendor data formats and the need to connect data from multiple sources makes scientist daily workflows challenging. This presentation will demonstrate how Allotrope Simple Model (ASM) standardization of chromatographic data at scale simplifies technical architecture to improve scalability and creates transformative opportunities for on demand data consumption (visualizations, feed AI/ML pipelines, etc.).