Our vision at Data Science Services & Solutions, DS3, is to Enable, Scale, and Operationalize R&D Data Science & AI capabilities to achieve Clinical Research and Development Speed, Efficiency, and Optimization. Learn more about our areas of responsibility, capabilities, strategic priorities, and collaborations. For more information, contact DS3 Head: Abi Velurethu. Artwork created using Stable Diffusion.

Given the considerable amount of data that is collected throughout clinical trials, it can be asked if this data can be used to improve our understanding of the trial, regardless of outcome. The radiomics pilot seeks to provide a toolkit that is capable of analyzing the different modalities of information found in clinical trials (genome, gene expression, biomarker, clinical information, and imaging information) to derive insights into why a patient may be responding, or why a clinical trial was unsuccessful. The current efforts have focused on clinical trials in oncology, while collaborating with the RED-ONC function. Overall, the project is in its pilot state, and demonstrating a proof of concept, while documenting many challenges and learnings with the development of such a toolkit. Potential directions that could expand this toolbox include deriving the factors that best predict why a patient would stay in a clinical trial, and providing assistance with patient selection of clinical trials. Other potential directions involve suggesting why a particular trial may have been unsuccessful and providing that information back to the drug development team.

With the new head of R&D Community Engagement, RDCE, a new set-up has also been introduced, which newly regulates the tasks and responsibilities of the business partners assigned to the functions, SPOCs, which stands for Single Point of Contact. What is now in scope and what is not? The poster provides information on this.

The R&D Digital Fluency Program (DFP) is supporting you with learning offerings to build up competencies and further skillsets around digital technologies to enable a competitive and digitally literate organization. At the Digital Fluency Program, we set ourselves apart by offering learning opportunities created by your peers who understand the unique challenges you face in your job. Our tailored approach focuses on real-life use cases, taking our training far from generic and providing you relevant and impactful skill development. In 2020 we, a small team of four people, started our journey with the DFP as part of the R&D Digital Roadmap. Since technologies are ever evolving and impacting our work environment it is more important than ever to foster a culture of continuous learning and enable future opportunities. Thus the DFP learning portfolio includes something for everyone, regardless of your level of knowledge: It offers general awareness programs to enhance basic digital literacy, self-learning pathways on data analysis & visualization and the ability to access online courses like Coursera via the Digital Curriculum. The team ensured the learning objective was met by piloting newly developed learning programs, making use of cross-functional development, and incorporated R&D specific components prior to a broader roll-out in the organization.

Morphological profiling with Cell Painting allows an unbiased characterization of cellular states by observing changes in cell morphology. Cell Painting images are information-rich and can guide the elucidation of mode of action, toxicity, and off-target effects, help drug repurposing, indicate differentiation states of iPSC lines, discover new biology and more. However, this technology is data-greedy and computationally complex and demanding. At DS&AI, we enable this technology at Bayer! To this end, we support large data generation campaigns internally and externally, develop novel machine learning algorithms, and create tools for efficient analysis and visualization of Cell Painting data. For more information please contact Paula Marin Zapata or Marc Osterland. Self-painted image modified using pictures from https://quizlet.com/420027050/animal-cell-diagram/ as starting models.

At Language of Life, LoL, we combine state-of-the-art machine learning models, bioinformatics and expert knowledge to understand and design large biomolecules (RNA, DNA, Proteins), helping deliver the best drug, faster and cheaper, to address our patients’ needs. Our objectives are, 1) Drug the undruggable 2) Target any modality 3) Efficient Dry lab – Wet lab loops 4) Understand Patient & Disease

Over the past two decades, the interdisciplinary predicTeam has established a prediction platform at Bayer Pharma R&D with the goal to generate state-of-the-art machine learning models for a variety of pharmacokinetic and physicochemical endpoints in early drug discovery. These tools are accessible to all scientists within the company and can be useful in assisting with the selection and design of novel leads, as well as the process of lead optimization. The predicTeam provides an all-inclusive package covering the data pipeline from experiment to application in projects. In close interaction with experimentalists, we select endpoints for model building that are relevant for drug discovery. We implement and maintain the infrastructure to retrieve and prepare the data and make it accessible as a data lake. For each endpoint, after fully exploring the matrix of data, molecule representations and algorithms, we implement the best-performing and most stable-models models in our internal research platform Pix. A highly automated infrastructure allows us to perform weekly retraining of the models to ensure that the novel chemical space of drug discovery projects is well embedded. We ensure close interaction (e.g. presentations, tutorials, teams channel) with the user base for optimal model use and direct feedback allowing for constant improvements. Finally, our Model Performance Report helps users to assess the applicability of each model to their specific project molecules.

The Applied Machine Learning Group (AML) strives for the application of a data-driven approach to drug research and development by developing and utilizing best-in-class machine learning (ML) models to optimize R&D processes. The implementation of ML algorithms accelerates drug discovery and a platform to monitor and self-train models enables portfolio view of our capabilities, allowing us to become more focused. We envision to work in a fast-paced setting, we expect to fail quickly and adapt swiftly, all while expanding your ML capabilities. We also believe in making the right choices not only for our immediate team but decisions to build solutions catered to broader R&D. The success at AML will be measured by bucketing the benefit of the developed capability into one or more of these three buckets, (1) Savings in time (2) Savings in cost (3) Generation of new insights