Data in Biotech

This week, we are pleased to have Stewart Fossceco, Head of Non-Clinical and Diagnostics Statistics at Zoetis and an expert in pharmaceutical manufacturing, join us on the Data in Biotech podcast. We sat down with Stewart to discuss implementing and improving Quality Assurance (QA) processes at every stage of biotech manufacturing, from optimizing assay design and minimizing variability in early drug development to scaling this up when moving to full production. Stewart talks from his experiences on the importance of experimental design, understanding variability data to inform business decisions, and the pitfalls of over measuring.  Along with host Ross Katz, Stewart discusses the value of statistical simulations in mapping out processes, identifying sources of variability, and what this looks like in practice. They also explore the importance of drug stability modeling and how to approach it to ensure product quality beyond the manufacturing process. Data in Biotech is a fortnightly podcast exploring how companies leverage data innovation in the life sciences. --- Chapter Markers [1:39] Stewart starts by giving an overview of his career in biotech manufacturing. [3:54] Stewart talks about optimizing processes to control product quality in the early stages of the drug development process. [7:27] Ross asks Stewart to speak more about how to optimize and minimize the variability of assays to increase confidence in clinical results. [12:11] Stewart explains the importance of understanding how assay variability influences results and how to handle this when making business decisions. [14:13] Ross and Stewart discuss the issue of assay variability in relation to regulatory scrutiny. [17:07] Stewart walks through the benefits of using statistical simulation tools to better understand how an assay performs. [19:49] Stewart highlights the importance of understanding at which stage sampling has the greatest impact on decreasing variability [22:09] Stewart answers the question of how monitoring processes change when moving to full production scale. [26:39] Stewart outlines stability modeling and the importance of stability programs in biotech manufacturing. [30:38] Stewart shares his views on the biggest challenges that biotech manufacturers face around data. --- Download our latest white paper on “Using Machine Learning to Implement Mid-Manufacture Quality Control in the Biotech Sector.” Visit this link: https://connect.corrdyn.com/biotech-ml

This week, we are pleased to welcome to the Data in Biotech podcast Brendan McCorkle, CEO of SciNote, a cloud-based ELN (Electronic Lab Notebook) with lab inventory, compliance, and team management tools. In this episode, we discuss how the priorities of ‘Research’ and ‘Development’ differ when it comes to the data they expect and how they use it, and how ELNs can work to support both functions by balancing structure and flexibility. We explore the challenges of developing an ELN that serves the needs and workflows of all stakeholders, making the wider business case for ELNs, and why, in the lab, paper and Excel need to be a thing of the past. Brendan is upfront about the data challenges faced by biotechs, which do not have one-vendor solutions. He emphasizes the importance of industry collaboration and software vendors’ role in following the principles of FAIR data. We also get his take on the future of ELNs and how they can leverage AI and ML. Data in Biotech is a fortnightly podcast exploring how companies leverage data innovation in the life sciences. Chapter Markers [1:40] Brendan gives a whistlestop tour of his career and the path to setting up SciNote. [4:20] Brendan discusses the principles of FAIR data and the challenges of adhering to them in the biotech industry. [6:15] Brendan talks about the need to balance flexibility and structure when collecting R&D data. [13:34] Brendan highlights the challenge of catering to diverse workflows, even within the same company. [16:05] Brendan emphasizes the importance of metadata and how vendors, like SciNote, can help collect it with flexible tools for data entry and post-processing. [18:59] Ross and Brendan discuss how to create an ELN that serves all stakeholders within the organization without imposing creativity constraints on research scientists. [21:57] Brendan highlights how benefits like improving loss reduction and efficiency form part of the business case for a tool like SciNote. [24:25] Brendan shares real-world examples of how companies integrate SciNote into their organizations and the need to work with other systems and software. [34:01] Ross asks for his advice to biotech companies considering implementing ELNs, particularly into their workflows. [39:10] Brendan gives his take on incorporating ML and AI within SciNote. --- Download our latest white paper on “Using Machine Learning to Implement Mid-Manufacture Quality Control in the Biotech Sector.” Visit this link: https://connect.corrdyn.com/biotech-ml

This week, we are pleased to be joined on the Data in Biotech Podcast by Pierre Salvy, who recently became the CTO at Cambrium, and his colleague Lucile Bonnin, Head of Research & Development at Cambrium.  As part of the Cambrium team behind NovaColl™, the first micro-molecular and skin-identical vegan collagen to market, Pierre and Lucile share their practical experiences of using AI to support protein design. We ask why Cambrium, as a molecular design organization, decided to focus on the cosmetics industry and dig into the factors that have driven its success. From developing a protein programming language to the challenges of collecting and utilizing lab data, Pierre and Lucile give a detailed look under the hood of a company using data and AI to accelerate its journey from start-up to scale-up. They also talk to host Ross Katz about the benefits of working as a cloud-native company from day zero, de-risking the process of scaling, and opportunities for new biomaterials.    Data in Biotech is a fortnightly podcast exploring how companies leverage data innovation in the life sciences.  Chapter Markers [1:34] Pierre and Lucile make a quick introduction and give an overview of Cambrium’s work using AI to design proteins with the aim of developing sustainable materials. [4:00] Lucile introduces NovaColl™, and Pierre elaborates on the process of bringing Cambrium’s first product to market. [7:37] Ross asks Pierre and Lucile to give an overview of the considerations and challenges of protein design. [11:01] Pierre and Lucile explain how Cambrium works with potential customers to design specific proteins that meet or exceed their expectations. [12:49] Ross and Pierre discuss how Cambrium approached developing the data systems it needed to explore the protein landscape and how the team optimized the lab set-up. [18:04] Pierre discusses the protein programming language developed at Cambrium. [21:24] Lucile and Pierre talk through the development of the data platform at Cambrium as the company has scaled and the value of being cloud-native. [24:12] Lucile and Pierre discuss how they approached designing the manufacturing process from scratch and how to reduce risk at every stage, especially while scaling up.   [31:44] The conversation moves to look at how Cambrium will use the processes and data platform developed with NovaColl™ to explore opportunities for the development of new biomaterials.  [34:42] Pierre gives advice on how start-ups can be smarter when selecting an area of focus. [36:27] Lucile emphasizes the importance of getting cross-organizational buy-in to ensure successful data capture.  [39:01] Pierre and Lucile recommend resources that may be of interest to listeners seeking more information on the topics covered.  --- Download our latest white paper on “Using Machine Learning to Implement Mid-Manufacture Quality Control in the Biotech Sector.” Visit this link: https://connect.corrdyn.com/biotech-ml

Exploring the importance of strong data foundations in biotech startups with Jacob Oppenheim. Discussing challenges faced with existing tools, limitations, and the role of AI in the biotech space. Insights on new modelling capabilities, building a team within a startup, and the role of consultancies in providing expertise.

Exploring OmicSoft's NGS analysis suite for in-depth exploration among datasets, challenges of acquiring NGS data, customer interactions, and the build vs buy dilemma in bioinformatics tools.

In this podcast, Wolfgang from Merck discusses BayBE, an open-source library for Bayesian optimization, its benefits, challenges, and the need for standardized data models. He shares insights on accelerating innovation, experimental design, and advice for scientists in the industry.

This week, we’re delighted to be joined by Annemie Ribbens, VP Science, Evidence and Trials at icometrix, a medical technology manufacturer that offers a portfolio of AI solutions to assist healthcare with various challenges in neurological disorders, such as brain trauma, strokes, dementia, and Alzheimer's disease. During this episode, Annemie opens up on icometrix’s mission in analyzing and treating neurological disorders, the work that went into developing the data infrastructure and the challenges they face when dealing with such large data sets. Annemie also goes on to discuss how machine learning will influence the application of precision medicine in biotech over the next five years and the goals that the company is looking to achieve in the future.  Data in Biotech is a fortnightly podcast exploring how companies leverage data innovation in life sciences. Chapter Markers: [1:14] Annemie provides us with a brief introduction into her background and what led her to pursue a career in this field. [4:17] Ross asks Annemie about icometrix’s portfolio and what differentiates it from the other tools in the market.  [7:36] Annemie explains how icometrix are helping physicians improve both their understanding and treatment of particular disorders.  [13:31] Annemie dives into the role that the public patient facing app plays and how the data that it gathers feeds the ecosystem.  [22:03] Annemie reveals how their partnerships work.  [28:11] Ross asks Annemie to provide some insights into how icometrix went about developing their data infrastructure. [31:23] Annemie shares the channels involved when processing and analysing large data sets.  [40:01] Annemie explains the methodology that enables icometrix to know what core areas to focus on. [43:07] Annemie reveals one of the projects that she is most proud of.  [45:00] Annemie gives us her thoughts on what the future holds for machine learning.  [47:33] Annemie explains where listeners can go to find out more information on icometrix.  --- If you’re a biotech company struggling to unlock a data challenge, CorrDyn can help. Whether you need to supplement existing technology teams with specialist expertise or launch a data program that lays the groundwork for future internal hires, you can partner with Corrdyn to unlock the potential of your business data - today. Visit connect.corrdyn.com/biotech to learn more.

Lana Denysyk, Head of RWD Assets at Novo Nordisk, discusses the use of real-world data in biotech. She explores the diverse data ecosystem in organizations and the importance of understanding which data type best suits specific research questions. Lana also highlights the role of RWD professionals in facilitating broader access to data and the challenges of navigating privacy regulations. The chapter descriptions cover the integration of real-world data into biotech organizations, managing resource constraints, standardized processes, privacy regulations, and developing a data strategy for pharmaceutical companies.

Cody Schiffer, Associate Director, Machine Learning at SMPA, discusses the construction and application of a knowledge graph in biopharma, integrating structured and unstructured data for tasks like literature searches. He explains how the knowledge graph was constructed and updated over time, incorporating unstructured data and applying information weighting. Cody shares lessons learned and the view on the build versus buy question.

This week, we’re delighted to be joined by Vera Mucaj, Chief Scientific Officer at Datavant, a data logistics company for healthcare. During the interview, Vera shares how the organization helps companies overcome issues of data fragmentation and availability, which ultimately leads to better decision-making in healthcare, the importance of connecting clinical trial data with real-world data to improve research outcomes and the challenges of semantic interoperability in data sets. Data in Biotech is a fortnightly podcast exploring how companies leverage data innovation in life sciences. Discussed: [1:06] Vera introduces herself, explains her role, and explains what drove her to join the scientific field. [4:00] Vera outlines the challenges that Datavant customers face in the acquisition and sharing of medical data. [8:49] Vera unpacks the role Datavant plays in enabling healthcare organizations to produce and consume data to answer important scientific questions. [12:21] Vera addresses the problem of security around data assets and how the company helps producers tackle the risk of the potential de-identification of sensitive patient data. [15:27] Vera explains the most common use cases of the Datavant platform and the value that can be unlocked by consumers of the platform. [22:10] Vera outlines Datavant's business model. [27:06] Vera talks about how Datavant is leveraging AI to generate, analyze and create imputations of data sets. [31:09] Vera discusses the regulatory and cultural barriers to making data available in healthcare. [38:53] Vera shares her thoughts on semantic interoperability between data sets. [44:31] Vera shares the resources that she follows to better understand the work of Datavant. --- If you’re a biotech company struggling to unlock a data challenge, CorrDyn can help. Whether you need to supplement existing technology teams with specialist expertise or launch a data program that lays the groundwork for future internal hires, you can partner with Corrdyn to unlock the potential of your business data - today.  Visit connect.corrdyn.com/biotech to learn more.