Quantitative analysis creates novel weighting scheme

Using quantitative analysis methods, scientists have created a weighting scheme to remove the noise from protein-protein interaction graphs.

The team of scientists were inspired to undertake the study, which is published in BMC Bioinformatics 2011, as developments in technology applied to biology, such as yeast-two-hybrid, phage display and mass spectrometry, have enabled the creation of detailed maps of protein interaction networks.

However, despite being a rich source of data, the findings are marred by noise.

As a result the scientists endeavoured to eliminate the noise inherent in interactome data in a two-step procedure using first novel and well established protein-protein interaction (PPI) weighting method and then applying clustering using established algorithms.

The team found that their experimental weighting method clearly improved the results of several clustering algorithms.

Furthermore, the novel weighting scheme outperformed other more established approaches to PPI graph weighting.