The Analysome is the term given to the collection of all analytes that can be measured, the technologies that can measure them, the limits of quantification of those technologies and the cost of obtaining those measurements. Such information – if curated, stored and accessed via an easy-to-use, globally-available, web-based catalogue – would be of value to the life science R&D community. The prototype demonstrator was made available at Thomson Reuters Cortellis Labs but is no longer online.
Semantically Enriched Scientific Literature (SESL)
Why should scientific information be handled differently from retail goods or travel products? The SESL (Semantically Enriched Scientific Literature) project imagined that scientists could look for information on disease-causing genes the same way they would search for a toaster or the cheapest fare for an upcoming vacation. Plug in criteria, and let an Amazon-com-like brokering service aggregate information from all available sources and display it for convenient browsing.
The service envisioned by the SESL team would “push” appropriate information to scientists based on a single query, rather than requiring scientists to “pull” information from often disconnected databases and literature sources possessing a multiplicity of different user interfaces.
The SESL team completed a public demonstrator for brokering gene data associated with Type 2 diabetes in 2011. The demonstrator included content from numerous bioinformatics databases (Uniprot, OMIM, and ArrayExpress) and five publishers (Nature Publishing Group, Oxford University Press, the Royal Society of Chemistry, UK Pubmed Central, and Elsevier). While a second phase of the project was envisioned, the team opted instead to transfer learnings from the project to other semantics-based projects, most notably the Innovative Medicines Initiative Open PHACTS consortium.
Project participants included:
- Wendy Filsell (project co-lead), Unilever
- Ian Harrow (project co-lead), Ian Harrow Consulting
- Roche, GSK, AstraZeneca, Pfizer, Unilever, EMBL-EBI, Nature Publishing Group, Royal Society of Chemistry, Oxford University Press, Elsevier
Electronic Laboratory Notebook (ELN)
By 2010, ELNs had become a fixture in most pharma labs, particularly in discovery-focused chemistry labs. The increasing virtualization of drug discovery coupled with the collaborative nature of pharmaceutical research challenged life science organizations to find ways to integrate data effectively across ELN platforms.
The Pistoia Alliance ELN project aimed to create a common standard for sharing information between and conducting searches across different ELNs.
Decrease total search time by ensuring scientists would not have to conduct duplicate searches across multiple, separate ELNs
Reduce the cost associated with transforming and reformatting data and building custom integrations
Enable organizations to extract more value from deploying ELNs in their organizations and partner sites.
Provide more definitive and useful shared knowledge repositories.
The project proceeded through two phases, during which use cases were developed by team members and an XML-based query standard was proposed. GGA, a specialist systems integrator, was retained to deploy a reference implementation of the standard in 2011. A lack of funding and critical mass on the project team, however, resulted in the termination of this activity in 2011.
Project participants included:
- Richard Bolton (project lead), GSK
- AstraZeneca, Bristol Myers Squibb, GSK, Novartis, Pfizer, Accelrys, CambridgeSoft, Edge, GGA
The common standard would:
- Decrease total search time by ensuring scientists would not have to conduct duplicate searches across multiple, separate ELNs
- Reduce the cost associated with transforming and reformatting data and building custom integrations
- Enable organizations to extract more value from deploying ELNs in their organizations and partner sites.
- Provide more definitive and useful shared knowledge repositories.
Automatic Structure Verification (ASV) from NMR Spectra
NMR analysis is a key quality control step in compound structure verification. Automatic structure verification (ASV) algorithms analyzing NMR spectra exist, but in general these algorithms are not sufficiently robust. Additionally, most software designers and vendors do not have sufficient NMR spectra/compound structure data to allow them to improve and test their algorithms.
The ASV effort aims to create a sufficiently large and diverse pool of NMR spectra and compound structure data to improve ASV algorithms.
The ASV team is currently seeking funding to enable the activity to become a fully-fledged project. Interested parties should contact the Executive Director for more information.
The project was closed in 2013 after funding didn’t materialise.
Project participants included:
- John Hollerton, project lead, GSK
The effort has the following goals:
- Build a database of at least 10,000 compounds from at least 10 companies to ensure size and diversity
- Provide some of the data for developers to improve algorithms and the rest of the data for pharma to use for testing
- Create better, more robust algorithms to eliminate this QC bottleneck and improve discovery cycles.
Vocabulary Services Initiative
The vocabulary services initiative aims to realize a new environment where project teams are able to use data from any partners with which they choose to engage and from any electronic systems they wish to interrogate. The formation of the VSI team was inspired by the Harvard Business Review’s selection of “the creation of agreed-upon standards for digitally representing drug assets” as one of the ten “breakthrough ideas” for 2010.
The VSI team published a paper in Drug Discovery Today reviewing the need for shared services around vocabularies. VSI is currently seeking funding to recruit a qualified consultancy to review vocabulary activities in life sciences R&D and identify potential project ideas.
The project ended in 2013 after funding failed to materialise.
Precompetitive development of open research vocabularies would benefit information producers and consumers by enabling them to:
- Share the development effort and cost of vocabulary production and maintenance
- Reduce redundancy and provide greater coverage of more concepts
- Better represent scientific domains and language differences by drawing on a wide body of global subject-matter experts
- Spend more time exploiting information and less time managing information
- Make outsourcing and collaborative business models more efficient
The App Strategy
Create a Pistoia Alliance “App Store” and associated governance to encourage innovation through the creation, storage, intercommunication, and dissemination of “apps” of appropriate quality and utility relevant to life science R&D.
Apps have become ubiquitous in consumer products and in many industrial settings, providing mobility and ease of use that have changed the way users interact with computers and data. Life science R&D, however, has been slow to embrace apps in a substantive way. The Pistoia Alliance’s app strategy aims to bring together the three necessary components to “appify” R&D:
- The technology components needed to produce useful R&D apps
- The scientific software engineers with the skill to assemble meaningful apps
- The life science researchers who want to use the finished products
By encouraging all parties to come together, the Pistoia Alliance will create a community where professionals can engage one another to map out solutions to problems with a high signal to noise ratio. The project will proceed in three phases:
- Launching an AppStore to collect together current apps relevant to life science R&D
- Enabling developers to submit apps directly to the Pistoia Alliance AppStore
- Constructed cloud-hosted service (e.g., data storage and calculation services) to support apps in the Pistoia Alliance AppStore
The project was drawn to a close in September 2014 as rapid technology progress in the marketplace had made many of the requirements redundant.
- Ingrid Akerblom (project lead), Merck
- Alex Clark (project analyst), Molecular Materials Informatics
- Roche, GSK, Bristol Myers Squibb, Collaborative Drug Discovery, Merck