Skip to main content
Premium Trial:

Request an Annual Quote

Columbia U Spinout Therasis IDs Genes Behind Brain Cancer Aggressiveness

NEW YORK (GenomeWeb News) – Scientists from drug-discovery company Therasis and Columbia University have identified a pair of genes that, when simultaneously activated, cause the most lethal form of glioblastoma.

The findings help validate Therasis' bioinformatics and systems biology approach to reconstructing maps of regulatory networks in specific cancer cell types, which could lead to the discovery of new biomarkers and therapeutic targets, a company official said this week.

The research, published last week in the advanced online version of Nature, was conducted by Andrea Califano, co-founder of Therasis and director of the Joint Centers for Systems Biology at the CU Medical Center; and Wei Keat Lim, head of computational systems biology at Therasis.

The scientists identified the genes by "reverse-engineering" a map of the molecular interactions that occur within human glioblastomas using computational methods and algorithms developed in Califano's lab.

Specifically, the team used an algorithm called ARACNe to reconstruct the cellular network that controls the behavior of the glioblastomas, and an algorithm called MARINa to identify the master regulators involved in the worst prognoses of the disease.

The scientists identified two genes – transcription factors called C/EBPβ and STAT3 – that had no prior association with brain cancer, but which played a synergistic role in determining the most aggressive properties of glioblastoma, including the ability to invade normal surrounding tissue and angiogenesis.

Califano and colleagues followed up their computational findings with gene expression studies, which demonstrated that co-expression of C/EBPβ and STAT3 were strongly correlated with increased mortality in glioblastoma patients. The scientists also confirmed the cellular interaction network and gene function in cell lines and mouse models, and showed that silencing the genes in human glioma cells transplanted in mice blocked their ability to form tumors.

"Many companies are touting themselves as systems biology companies … but the reality is that right now [no one] can produce experimentally validated, fully integrated models of regulation for specific cells," Califano told GenomeWeb Daily News this week. "What we call pathways right now are essentially small snippets of the regulatory machinery that have been identified across all sorts of different cells."

Moreover, many of these pathways are identified using ex vivo assays and don't truly reflect the actual interactions that are occurring in cells, Califano said.

"The paper really illustrated that … we can reconstruct to a very high degree of accuracy and completeness a repertoire of molecular interactions in the cell," Califano said. "There are literally hundreds of thousands of interactions of a regulatory nature that are inferred de novo. The paper showed that you can build a map, validate it, and show that a large number – 80 to 90 percent – of the interactions we predicted in fact implemented themselves."

Califano added that such a map will allow scientists to ask "very pointed questions" about specific signatures and molecular regulators of particular cancer types.

New York-based Therasis, which spun out of Columbia in 2007, launched its oncology drug-discovery engine, dubbed the Therasis Filter, earlier this month, buoyed by a $12 million Series A financing round led by Tilocor Life Science.

Therasis said that it plans to use its systems biology approach to screen its own library of compounds for potential therapeutic candidates for treating lymphoid malignancies and breast cancer; and to develop "long-term" partnerships with pharmaceutical companies to co-develop therapeutic compounds of interest, Califano said.

Therasis has taken a license to a patent application owned by Columbia covering methods of using several different algorithms to reconstruct and interrogate molecular networks in cells for the purposes of identifying genes that are responsible for specific cellular traits.

Use of these algorithms is freely available to the academic research community, Califano said, but Therasis has an exclusive license from Columbia to use them for commercial purposes. In exchange for the license, Columbia has a small, undisclosed equity stake in Therasis.

The Scan

Shape of Them All

According to BBC News, researchers have developed a protein structure database that includes much of the human proteome.

For Flu and More

The Wall Street Journal reports that several vaccine developers are working on mRNA-based vaccines for influenza.

To Boost Women

China's Ministry of Science and Technology aims to boost the number of female researchers through a new policy, reports the South China Morning Post.

Science Papers Describe Approach to Predict Chemotherapeutic Response, Role of Transcriptional Noise

In Science this week: neural network to predict chemotherapeutic response in cancer patients, and more.