Collaboration with Toronto University (2007)

The University of Toronto is Canada's largest university, recognized as a global leader in research and teaching. The University of Toronto is a distinguished faculty, an institutional record of groundbreaking scholarship and wealth of innovative academic opportunities, which continually attract outstanding students and academics from around the world.

Scientists from Toronto University

The scientists who developed the Q5 algorithm are:

Ryan Lilien, MD, PhD (Toronto University, Canada)
Hany Farid, Professor, PhD (Dartmouth College, USA)
Bruce R. Donald, Professor, PhD (Duke University, USA)

Q5

MedicWave collaborates with a research group from Toronto University in Canada. During our collaboration we received a unique algorithm called Q5. This algorithm will be part of MedicWave's PDS™ software.

Q5 is an algorithm for probabilistic classification of healthy vs. disease whole serum samples using mass spectrometry. The algorithm employs Principal Components Analysis (PCA) followed by Linear Discriminant Analysis (LDA) on whole spectrum Surface Enhanced Laser Desorption / Ionization Time of Flight (SELDI-TOF) Mass Spectrometry (MS) data, and is demonstrated on four real datasets from complete, complex SELDI spectra of human blood serum. Q5 is a closed-form, exact solution to the problem of classification of complete mass spectra of a complex protein mixture. Q5 employs a probabilistic classification algorithm built upon a dimension-reduced linear discriminant analysis. Q5 is computationally efficient; it is non-iterative and computes the optimal linear discriminant using closed-form equations. The probabilistic classification method achieves excellent performance. Q5 achieved a sensitivity, specificity, and positive predictive value above 97% on three ovarian cancer datasets and one prostate cancer dataset. The Q5 method outperforms previous full-spectrum complex sample spectral classification techniques, and can provide clues as to the molecular identities of differentially-expressed proteins and peptides.

Click here to learn more about Q5, and click here to read the algorithm description.

Publications concerning Q5

"Probabilistic disease classification of expression-dependent proteomic data from mass spectrometry of human serum", Lilien, R.H., Farid, H., Donald, B.R., JOURNAL OF COMPUTATIONAL BIOLOGY 10 (6): 2003

Open for collaboration

MedicWave develops state-of-the-art algorithms for data mining and machine learning in collaboration with renowned researcher from around the world. These algorithms are implemented into our bioinformatics products and used for differential proteomics studies, quantification and validation of regulated proteins, and the discovery of biomarkers.

Collaboration: Read more...