Last week, Large Scale Biology Corporation completed the first version of its proprietary human proteome database, the Human Protein Index v1.0.
The company carried out the work at its Rockville, Md., proteomics subsidiary, Large Scale Proteomics, using its ProGex proteomics platform to quantify measurements of the protein components in human tissues and compile them in a database. The company declined to elaborate on the ProGex technology.
LSBC examined a set of 157 tissue databases, which yielded a total of 115,693 proteins. The company estimates that HPI v1.0 currently covers the protein products of 18,000 human genes.
Leigh Anderson, president of Large Scale Proteomics, said that HPI measures “how much of those proteins is really there,” which sets it apart from other proteomics databases, such as Proteome’s BioKnowledge Library. “Proteome has databases of the annotation of known proteins as derived from scientific literature. It doesn’t include the actual measurements on samples,” he said.
Anderson noted that Proteome’s data is “completely synergistic” with the data contained in the HPI. “It provides information about the proteins that we’re detecting in all these tissues.”
While the company initially intends to mine the HPI for diagnostic and other markers internally, Anderson said that it would also leverage it in collaborations aimed at finding additional therapeutic targets or disease markers through proteomics.
LSBC is in ongoing discussions on several such collaborations.
Anderson said that the company hasn’t ruled out offering its information as a database product down the road, however.
LSBC currently uses the HPI as the core of its Molecular Anatomy and Pathology database, which compares diseased and normal tissues to find pathology-related proteins. Anderson said that the “low-hanging fruit” in this effort is deriving a series of proteins that would be the most effective tissue damage diagnostic markers. The company is continuously filing IP in this area, Anderson said.
HPI v1.0 is considered complete because “it’s the first assembled survey across essentially all of the major human tissues,” Anderson said.
Future versions of the HPI will include progressively larger fractions of the proteome, Anderson said, although there’s no specific timeframe for future releases.
We’ll choose what to announce as the second version when we feel like we’ve reached the next major watershed.” He said. “It sounds facetious, but you actually do know when you get there.”
“This is somewhat akin to software release versions,” he said. “And the main issue there is to basically work with release versions that are synchronized with what the users really need from them.”
A paper describing the completion of the database is scheduled to appear in the first issue of the journal Proteomics, published by Wiley-VCH, on January 18.
— Bernadette Toner