Compugen and Sigma-Genosys this week announced the release of an expanded human OligoLibrary, a collection of some 29,000 oligonucleotide probes representing a similar number of genes from both public and proprietary provenance.
Tel Aviv-based Compugen has designed the library of oligos, which are manufactured by Sigma of The Woodlands, Texas.
In this effort, Compugen has marshaled its ten years of experience in life sciences informatics toward oligonucleotide design. Using its Leads technology for transcriptome prediction, the company says it has an oligo design process that incorporates alternative splicing models, reduces redundancy in gene representation, minimizes cross-homology for higher specificity, and improves sequence quality by avoiding sequencing errors and polymorphic sites.
The list price of the new oligo set is $119,000 for a quantity of 1 nanomole, with discounts to existing or early-access customers.
The library of human-gene oligos follows the completion of the human genome in April, but is based on more than two years of research and development, said Albine Martin, vice president of business development for Compugen.
“We initiated this concept of developing probes for whole-genome analysis two to three years ago,” she said. “[For this release], we saw a need to update the representation of the human genome — as we understand it today.”
Researchers are ready for the expanded library and 10,000 new probes, and genes.
“I think this will make a difference,” said Lance Miller, a former National Cancer Institute scientist who now heads microarray and expression genomics at the Genome Institute of Singapore. “We have been using their first release human array library of 19,000 elements and it will be helpful to have more genes to represent on arrays. The majority of the work that we do will benefit.”
The Genome Institute of Singapore performs much of its microarray analysis on a home-brew platform, in addition to studies that are conducted on the Affymetrix platform, Miller said. His lab prints 250 microarrays at a run and some 6,000 microarrays per year.
He said the lab chooses to do the majority of gene-expression research using self-spotting arrays for clear economic reasons.
“To do what we do [solely] on the Affymetrix platform would be prohibitive in terms of cost,” he said. “We can make arrays [based on Compugen’s probes] for about $10 a chip. That’s an order of magnitude difference.”
Based on quantitative PCR checks, the oligos are accurate, he said.
“We find these oligos accurately report the transcript levels within our studies,” Miller said. “Whenever we find oligos that are giving us expression measurements that show differential expression, we can confirm those by quantitative PCR, to a large extent. We look at the array data in terms of the biological composite picture, what is going on inside the cell and system we are using. We often see many, many genes representing biological pathways coming up, again, confirmation that the probes are performing well.”
Miller said he would like industry to help him design probes that are specific for certain transcripts.
“We put probes on the array that represent the gene,” Miller said. “That gene is actually represented by multiple transcripts, in most cases, yet we don’t have probes that discriminate them. I think in some context, you can’t design oligos to discriminate between the transcripts because they don’t exist. But, in other cases, you can discriminate between transcripts to some degree. It takes a clear understanding of how transcripts vary for a given gene, and that’s in the context of a biological space, which is vast. Some tissues might only have one variant and other tissues might have multiple splice variants expressed at a given time.
“I’ve always liked the Compugen approach, the Leads clustering system for pulling out of the databases all of the transcripts and orienting them in such a way, and [their] GeneCarta helps facilitate the process. You can pull up all of the transcript information representing a given gene — at least what is in the databases. Then using that platform, one can think about designing oligos that recognize a particular transcript.”
“It’s a daunting task, to take on the whole genome,” he said.
Martin said that is a possibility with this new library.
“A collection like this allows the user to do a great deal of customization,” she said.