NEW YORK (GenomeWeb) – Bioinformatics firm Repositive has kicked off a collaboration with AstraZeneca to put together a pre-competitive resource on the Repositive platform that will provide pharmaceutical companies and drug discovery organizations with access to patient-derived xenographs (PDX) and associated molecular data for use in oncology research.
Repositive is also preparing for the commercial launch of its platform on Sept. 1, company CEO Fiona Nielsen told GenomeWeb this week. The company has been beta testing its platform for about a year and getting feedback from the community about what features to include to improve the resource. In February, the company received about $721,000 in seed funding for platform development and to prepare for the fall launch. This is in addition to about $446,000 raised last year to support the development of the platform.
Repositive acts as an intermediary between data providers and researchers. Within its platform, researchers can find out what datasets are available to them and filter through a number of possible options to select the sources that best suit their purposes. The company does not host datasets internally, rather links to the repositories where the data is located within its platform so that users can access and download datasets directly from the source if they are public, submit access applications for restricted datasets in repositories such as dbGAP, or purchase licenses in the case of commercial databases.
Researchers can search through available datasets and identify which ones are most relevant for their projects and studies. The company has sourced and indexed these datasets using metadata provided by developers of various public, semi-public, and proprietary data repositories. There are also mechanisms for querying genomic datasets containing sensitive information and extracting basic statistics from them without compromising the privacy of the individual genome contributors.
When it launches in the fall, the platform will include information on lots of different datasets, Nielsen said. In addition to the PDX data, other datasets currently accessible via the Repositive platform includes ones from repositories such as Array Express, dbGaP, and the Gene Ontology. Earlier this year, Repositive announced a partnership with Xpressomics to give Repositive users access to that firm's gene expression data repositories. The company also signed a partnership last year with GigaScience linking its platform to a range of biomedical science research resources.
Moving forward, Repositive will also explore additional collaborations with its user base focused around specific datasets similar to the current collaboration with AstraZeneca around PDX data, Nielsen said. The current consortia project was inspired by an idea submitted by AstraZeneca to the Pistoia Alliance, a global, not-for-profit alliance of life science companies, vendors, publishers, and academic groups dedicated to improving global life sciences research and development. The idea came through Pistoia's IP3 platform, which provides a forum for launching and establishing pre-competitive collaborative projects. Projects launched on the platform include the Ontologies Mapping project, launched last year, which aims to develop better tools and establish best practices for ontologies used in life sciences research and development.
"When we met AstraZeneca and talked to them about our expertise in finding and accessing data, they got really excited because this project is all about finding and accessing the right data for the PDX models," Nielsen told GenomeWeb. Previously a bioinformatics team within AstraZeneca maintained a list of relevant PDX models and vendors which resulted in significant overhead and took valuable time away from analyzing data.
Under the terms of the partnership, Repositive will now maintain AstraZeneca's list on its platform as well as provide access to information on other PDX models that Repositive has gleaned from other sources including Jackson Laboratories and Horizon Discovery. In addition to the models, PDX vendors provide data on their genotype and genetic makeup as well as whole-genome sequence and expression data in some cases.
"We are indexing the descriptions of data and the different types of data so that [researchers] can ask specific questions like 'where do I find a PDX model that has mutations in p53, for example, or where do I find a PDX model that has these certain copy number variants," Nielson explained. Also, "we link to the vendor of the PDX model [so that] the pharma company [can] then approach that vendor and say 'we are interested in your model, can you run this experiment for us or can you source us [a] model treated in this way." All of the information and the search functionality will be accessible within the Repositive platform when it officially goes to market in the fall.
Repositive and AstraZeneca are seeking additional industrial and academic partners to provide improved data and contribute new tools for oncology research. The goal of this initiative is to create a public shared resource that AstraZeneca and other pharma and oncology research organizations can use for their drug-discovery projects. Such a resource would free them to focus on their actual research interests rather than on maintaining a database and keeping it up to date, Nielson said, and it would also reduce duplication of effort since its quite likely that other pharma companies maintain their own internal repositories of PDX resources as AstraZeneca has done to date.
They are inviting Pistoia Alliance members and other interested organizations to join the collaborative project as partners in developing the PDX resource. "It's all about making the process of the drug discovery pipeline more efficient," Nielsen said. "This resource gives you fast access to find the specific data that you are after. For pharma companies, this has a huge impact because if you can find the right data fast, that means you can reach results fast, get rid of false positives faster. These are things that can shorten the time for your drug to market."
As an incentive to join, PDX consortia partners will benefit from early access to the resource for about a year, after which it will then be made more broadly available to the non-member oncology community, Nielsen said. Furthermore, consortia members will have the opportunity to give their input on what features and datasets should be included in the resource as it is developed moving forward, she said. There will also be opportunities to extend the resource to include information on immuno-competent tumor models, according to Repositive. To sign up as a consortia partner, prospective partners must express their interest with Repositive before Sept. 30.