Following a year of evaluation, Huntington’s disease research organization CHDI has selected the Ingenuity Pathway Analysis system to serve as the foundation for a disease-specific research informatics platform it is developing for internal use.
Under the terms of the multi-year agreement, for which financial terms were not disclosed, Ingenuity Systems will create a customized version of IPA that will serve as the central access point for CHDI’s proprietary Huntington’s-specific data as well as for disease-specific information that Ingenuity’s curators will extract from the scientific literature.
“Ingenuity is going to curate content that we provide in addition to expanding the scope of the publicly available content that they curate in order to embrace additional material that they would not otherwise have curated for the general IPA release,” Richard Morse, director of computational and structural biology at CHDI, told BioInform this week.
The agreement is the first disease-specific curation partnership that Ingenuity has announced, but company officials said that it is not its first such deal. Rather, it is just the first it has had permission to disclose because its previous projects have been with pharmaceutical customers who requested confidentiality.
“We’ve actually been doing disease-specific curation projects for a long time,” said Heidi Bullock, director of marketing at Ingenuity. “It’s something that we’re seeing happen more and more because that’s where people really have their competitive edge — it’s the area that they are particularly focused on.”
Other pathway informatics companies are pursuing a similar path. For example, GeneGo recently embarked upon a project to develop an oncology-specific version of its MetaCore and MetaDrug platforms, a project that it began after earlier launching a version of its software focused on cystic fibrosis [BioInform 06-20-08].
Bullock said that the CHDI agreement marks the first disclosure of another trend the company has seen among its pharmaceutical customers — that of using IPA as a knowledge-management and -integration platform that provides access to internal and external data.
Ingenuity scientist Megan Laurance said that there are “at least two large pharma companies who have completely embraced the full knowledge-management approach using the Ingenuity platform,” though she could not disclose the names of these customers.
She added that knowledge management is a “longstanding issue for academic researchers and industry alike. “Our customers talk about it all the time,” she said. “There’s incredible internal knowledge that either doesn’t get integrated because someone is working on it and it’s buried in their notebooks, or it’s a list of genes in their mind,” she said.
Laurance said that using IPA as an integration platform can be “a real step towards a practical solution for making sure that that knowledge is mined and is used in a very practical way,” particularly for information “that might not necessarily ever get published in the peer-reviewed literature, but is critical to generating testable hypotheses.”
CHDI is the successor to the Cure Huntington's Disease Initiative, a program founded by the Hereditary Disease Foundation that was later spun off and renamed CHDI Inc.
Morse, who joined CHDI in 2005, is no stranger to taking a pathway-centric approach to data analysis. Prior to joining the organization, he worked as a computational biologist at Protein Pathways, where he developed pathway informatics software to support target identification.
Morse said that CHDI was not using IPA prior to this agreement, and that it had no legacy integration system to help its researchers access disparate data.
“We’ve actually been doing disease-specific curation projects for a long time.”
“This decision came out of a process of evaluation that we initiated over a year ago in order to help us to assemble the appropriate tools and technologies in order to enable our discovery biology target validation efforts,” he said.
The key criteria in choosing the platform, he said, was that it “had to be friendly to the end-user discovery biologists because they are the principal consumers.”
Morse said he realized quickly that “turning biologists loose on the raw tools available on the web — all the databases — was just not going to work. They needed one single point of entry, not having to go across multiple databases in order to pull information together.”
He said that he believes a pathway-based approach to data integration will enable CHDI researchers to “find unanticipated connections between putative targets that we can exploit for therapeutic purposes.”
He added, however, that the IPA system is not intended to replace other bioinformatics tools that CHDI researchers may want to use, but rather to complement them.
“We like to think that for the discovery biologists who are not informaticists that this would be the starting-off point, because of the way all of the knowledge from a lot of different sources has been centralized,” he said.
After that, he said, “as their questions become more refined, they can partner with a [CHDI] staff expert in order to go that extra distance if need be to find information that may not be right now captured in the Ingenuity software.”
Morse said he expects that the custom curated content will be available in the system in six to nine months.
While proprietary content will be accessible only to the end users of CHDI’s version of IPA, any information that Ingenuity curates through the public literature will eventually be available through the general release of the system, Laurance said.
She added that one of the “unique” aspects of the partnership “is that we are able to create proprietary instances of IPA and Ingenuity technology.”
Stretching the Ontology
The cornerstone of Ingenuity’s business since it was founded nearly a decade ago has been an internally developed ontology and a curation process that Laurance described as a “content-acquisition engine that we can really just point towards any scientific area.”
As a result, she said, the company doesn’t have to modify its curation process for a disease-specific project like the CHDI partnership. She noted, however, that the key to success for the project will be working closely with the CHDI researchers “to understand what level of detail is critical to capture from this content.”
Laurance said that based on the company’s previous experience with disease-specific curation, “it’s pretty unusual for us to find relationships and content that we don’t already have some type of structure for in the ontology that will handle it. So the real work is … making sure we’re capturing exactly what it is that the end users need to understand.”
She added that the company has developed its ontology in a way that enables it to evolve as necessary.
“It’s ready for new concepts,” she said. “So, if in the process of curating all this content, we run across a disease term, or a gene, or an miRNA, or something that is new to us, or new to our ontology, or new to science, we have easy processes for creating those terms and mapping them into the proper place in the ontology and creating all of the relationships that would occur between that new concept and all the existing concepts in the ontology.”
While acknowledging that other biological ontology projects have created specific ontologies for diseases and species, Laurance said that approach isn’t necessary under Ingenuity’s model.
“We certainly have ways of partitioning out that content very specifically, and we have different branches of the ontology, absolutely, but we don’t have a disease-specific ontology, so to speak,” she said.
“What we prefer to do is just develop queries that can, if necessary, set aside that content and present it into a disease-specific application like IPA.”