GOVERNMENT
Massively and CU Enter Into Relationship for Proteomics
Combining innovative proteomics application with Massively Parallel Technologies’ high-performance software technology holds the potential to transform speed and accuracy of disease biomarker discovery. -- Massively Parallel Technologies, Inc. (Massively), a provider of on-demand high-performance computing (HPC), announces it has entered into a cooperative agreement with the University of Colorado for the development and commercialization of an innovative breakthrough in proteomics analysis. 
Working with Mark Duncan, Ph.D., professor at the University of Colorado at Denver and Health Sciences Center (UCDHSC), departments of pediatrics, medicine and cellular and departmental biology and director of Proteomics Shared Resources, and his team, the two groups will apply Massively’s patented parallel processing software technology to a protein identification software application developed by UCDHSC. The novel analysis software is used to identify proteins from tandem mass spectrometry. UCDHSC’s unique approach will enable faster and more accurate results than what is currently possible but requires tremendous compute power to process the massive amounts of data in a practical timeframe. The compute scaling efficiencies and increase in performance in time-to-answer made possible by Massively’s technology will take the UCDHSC application to the next level of performance and usability, required to break through existing data analysis bottlenecks in the field of proteomics. Proteomics aims to define all proteins that are expressed by a cell, organ or organism. Identifying and fully characterizing the human proteome holds enormous potential to aid in clinical medicine and to enhance our understanding of human biology. The use of biomarkers for early disease diagnosis, better disease classification, predicting the outcome and response to a given therapy, and identifying new and high-quality targets for therapeutic intervention are just some of the potential benefits arising through proteomics. “We have invested a tremendous amount of resources to develop and validate the science behind this application. As a result, our application delivers high sequence coverage while reducing false positives and negatives in comparison to existing applications,” Duncan said. “The reality for us, however, is that without the high-performance computing resources, expertise and innovation that Massively brings to the table, this application would not be of practical use due to its computational demands.” Once proven, Massively will have two options to bring the solution to market and make it available to researchers world-wide. The first is via Massively’s Biotech Virtual Power Center – an Internet accessible, on-demand suite of software applications delivering supercomputing performance at PC pricing. The second option is to license the solution to the biotechnology and pharmaceutical companies. In addition to the initial protein identification application, Massively and UCDHSC are working together to identify and parallelize other key applications that promise to break through the data analysis bottleneck currently stifling the field of proteomics. “We envision a comprehensive suite of proteomics applications that will enable breakthroughs in everything from targeted disease treatment to personalized medicine,” said Scott Smith, CEO of Massively. “The cooperative agreement between UCDHSC and Massively along with our first National Institute of Health grant application is just the beginning of what we view as a long-term and mutually beneficial collaboration.” Massively’s patented and patent-pending HOWARD parallel processing software technology delivers the best price-performance value through unprecedented compute scaling efficiencies. Massively’s compute and cross-communication methods keep communication overhead growth rates, which determine maximum scaling efficiencies, to near zero as processing nodes are increased, even out to thousands of compute nodes so that the maximum power of each computer is actualized. Massively’s solutions can be made available through its Internet accessible, on-demand supercomputing Virtual Power Centers or on-premise installations.
Working with Mark Duncan, Ph.D., professor at the University of Colorado at Denver and Health Sciences Center (UCDHSC), departments of pediatrics, medicine and cellular and departmental biology and director of Proteomics Shared Resources, and his team, the two groups will apply Massively’s patented parallel processing software technology to a protein identification software application developed by UCDHSC. The novel analysis software is used to identify proteins from tandem mass spectrometry. UCDHSC’s unique approach will enable faster and more accurate results than what is currently possible but requires tremendous compute power to process the massive amounts of data in a practical timeframe. The compute scaling efficiencies and increase in performance in time-to-answer made possible by Massively’s technology will take the UCDHSC application to the next level of performance and usability, required to break through existing data analysis bottlenecks in the field of proteomics. Proteomics aims to define all proteins that are expressed by a cell, organ or organism. Identifying and fully characterizing the human proteome holds enormous potential to aid in clinical medicine and to enhance our understanding of human biology. The use of biomarkers for early disease diagnosis, better disease classification, predicting the outcome and response to a given therapy, and identifying new and high-quality targets for therapeutic intervention are just some of the potential benefits arising through proteomics. “We have invested a tremendous amount of resources to develop and validate the science behind this application. As a result, our application delivers high sequence coverage while reducing false positives and negatives in comparison to existing applications,” Duncan said. “The reality for us, however, is that without the high-performance computing resources, expertise and innovation that Massively brings to the table, this application would not be of practical use due to its computational demands.” Once proven, Massively will have two options to bring the solution to market and make it available to researchers world-wide. The first is via Massively’s Biotech Virtual Power Center – an Internet accessible, on-demand suite of software applications delivering supercomputing performance at PC pricing. The second option is to license the solution to the biotechnology and pharmaceutical companies. In addition to the initial protein identification application, Massively and UCDHSC are working together to identify and parallelize other key applications that promise to break through the data analysis bottleneck currently stifling the field of proteomics. “We envision a comprehensive suite of proteomics applications that will enable breakthroughs in everything from targeted disease treatment to personalized medicine,” said Scott Smith, CEO of Massively. “The cooperative agreement between UCDHSC and Massively along with our first National Institute of Health grant application is just the beginning of what we view as a long-term and mutually beneficial collaboration.” Massively’s patented and patent-pending HOWARD parallel processing software technology delivers the best price-performance value through unprecedented compute scaling efficiencies. Massively’s compute and cross-communication methods keep communication overhead growth rates, which determine maximum scaling efficiencies, to near zero as processing nodes are increased, even out to thousands of compute nodes so that the maximum power of each computer is actualized. Massively’s solutions can be made available through its Internet accessible, on-demand supercomputing Virtual Power Centers or on-premise installations.