PALO ALTO, Calif.--A week after announcing a new business plan that includes licensing components of its in-house technology engine to life sciences companies that seek to determine protein function, Molecular Applications Group said Monsanto will be its first such customer. Monsanto scientists will employ three of Molecular Applications' proprietary algorithms to identify, select, and prioritize targets for agricultural and pharmaceutical applications.
The three complementary techniques rely on hidden Markov models, protein threading, and phylogenetic analysis, according to Molecular Applications. The core technology is captured in three proprietary components that Molecular Applications calls SHMMs, ATHOS, and BETE. Paul Thomas, executive director of research for Molecular Applications, explained, "In combination, these algorithms produce a discovery system that greatly increases the reach of remote homologue detection, as well as the ability to classify novel genes in the context of related sequences." Thomas added that the highly scalable nature of the system will enable Monsanto to process large amounts of data rapidly .
Molecular Applications said the technology improves the prediction of gene and protein function and draws upon the company's expertise in protein sequence and structure analysis. "Determining the function of novel proteins continues to be one of the most difficult and time-consuming challenges we face, yet it is an essential step in efficiently identifying, prioritizing, and selecting targets," said Paul Spence, head of biotechnology for Monsanto's Searle division.
Monsanto will use the methods for associating target genes with genes that perform similar functions, even when there is minimal sequence similarity. In addition, automated subfamily classification of target sequences will be demonstrated on a large-scale basis, Molecular Applications claimed, saying that the project addresses the fact that new approaches are urgently needed to maximize the value of genomics information.