Output details

Hierarchical information clustering by means of topologically embedded graphs.

<13>Developing algorithms capable of extracting information out of complex datasets is a major current priority in science and industry. We propose a radically innovative approach that uses networks to filter out redundancy and dependency in signals, extracting meaning. This method is of general applicability: we discovered unknown links between cancer-types and gene-clones; we uncovered hierarchical organization in human brain activity (with clinicians at Imperial College London); we identified risk distribution in markets (with financial industry experts). Application to drug-redirection (with biotech company, mondoBiotech), produced groundbreaking results yielding to two new drugs (clinical trials in progress, patent pending).