Pistoia Alliance Project Digitizes Sharing of Analytical Methods Data To Support Machine Learning and Boost Cyber Resilience

The Pistoia Alliance is announcing that it has successfully completed a proof of concept (PoC) on the digital transfer via cloud of analytical High Performance Liquid Chromatography (HPLC) methods, proving it is possible to move analytical methods securely between two different vendors and Chromatography Data Systems (CDS) vendors with ease. This PoC delivers on a series of critical improvements to methods transfer that eliminate the manual keying of data, reduce risk, steps, and error while improving overall flexibility. The PoC is part of the Methods Hub Project, and has been successfully tested in labs at Merck and GSK, where there has been an effective transfer of HPLC information between different systems. The Alliance is now calling for broader participation from the pharma industry, as well as instrument and CDS vendors to support and fund future stages of the project, including identifying new use cases and further development of the cloud platform that enables the method transfer.

Greater digitalization of paper-based processes against standardized ontologies will help overcome the major industry barriers of reproducibility and interoperability, as well as removing opportunity for human error. This sets the stage for future data science technologies, such as AI and machine-learning, by ensuring data is machine-readable from the point of its creation. The project is also a step towards greater cyber-resiliency in the industry by providing a centralized place to store methods and promote the secure exchange of data. The Pistoia Alliance collaborated with the Allotrope Foundation, using the Allotrope Framework technology stack to ensure interoperability, as well as ZONTAL, which provided the virtual cloud where method information is stored according to FAIR principles. GSK, Merck and Agilent provided support and funding.

“This collaborative project is a critical first step in showing that it’s possible to digitally transfer chromatographic data between our developmental partners including methods, sequences, and results across different CDS platforms,” commented Ken Wells, Scientific Director and GSK Fellow. “That’s not possible without centralizing and standardizing experiment descriptions against a common data model – FAIR data is at the core of successful digitization in the industry.”

“Methods management – from storing to sharing, is one of the biggest problems our members face. When information is stored locally on a PC or an ELN that is not backed up, a single cyberattack can wipe them out instantly. Having a centralized storage of methods in a vendor-neutral format not only protects these notes, it makes them searchable and avoids unnecessary duplicate methods creation,” says Dr. Birthe Nielsen, Project Manager for Methods Hub. “In the future, we hope to link both analytical methods and results data in a single platform, which would help significantly in data analytics by enabling information from multiple vendors to be pooled and visualized simultaneously. These insights would never be possible without the groundwork this project is laying, including making data machine-readable.”

Companies frequently share methods both internally, and externally with a Contract Research Organization (CROs). This is a particularly important use case as clinical development is critical in developing new therapeutics. Consistent methods must be employed across organizations to facilitate regulatory approvals, a process which often holds back drug development. Often methods are simply shared over email or physically posted, with no centralized repository that scientists use to provide consistent access. CROs must re-establish and revalidate methods, which is labor-intensive, open to human interpretation or error, and ultimately worsens the reproducibility crisis. Improved methods exchange will create better flexibility, reproducibility, efficiency, and even improve security.

“This solution will increase efficiency and quality in method management and laboratory automation and unlock the data for advanced analytics applications scientists have long been waiting for. We cannot underestimate the impact this will have for pharmaceutical companies and their partners,” explains Wolfgang Colsman CEO, ZONTAL Inc. “HPLC method parameters are specific to both the Chromatography Data System and the HPLC Instrument Vendor and Model and have not been easily accessible in the past. This will be the first time method parameters and raw data are available in an interoperable data format across a variety of software and hardware systems commonly used in the industry.”

The Pistoia Alliance is looking for pharmaceutical companies, instrument and CDS vendors, and service organizations to come forward and participate in the next phase of the project. In this next phase, the Alliance plans to extend the platform’s functionality to connect analytical methods with results data, which would be an industry first. The Alliance aims to make Methods Hub an integral part of the system infrastructure in every analytical lab. It will do this by adding support for columns and additional hardware and other analytical techniques, such as mass spectrometry and nuclear magnetic resonance spectroscopy (NMR).