VIStA -- unsuperVIsed Shuffling Approach


INVENTORS: Jorg Menche; Albert-Laszlo Barabasi



Conventional patient clustering approaches rely on prior clinical characteristics. Such methods cluster patients using pair-wise similarity expression patterns.  As a result, these existing approaches allow for a biased division of patients, which at times is highly undesirable. This invention discloses a novel approach (unsupervised shuffling approach) that uses an algorithm to identify subgroup of patients with maximal difference in gene expression patterns, capable of overcoming prior-art limitations and/or unmet needs.

Value Proposition

The approach:

•Effectively divides a set of patients into different groups with a large number of differentially expressed genes

•Further identifies/pinpoints various clinical characteristics that differ significantly between groups

•Avoids a need for prior clinical patient information, enabling a complete unbiased assessment

•Avoids use of a predefined division of patients into case and control as required for conventional approaches

•Would be commercially useful for the following applications:

oIdentification of novel disease biomarkers

oIdentification of subtypes of complex diseases

oIndividualized diagnosis and treatment for personal medicine


Intellectual Property status

Provisional Application No. 61/843,682


Development stage


License status

Available for license


Patent Information:
For Information, Contact:
Mark Saulich
Associate Director of Commercialization
Northeastern University
Albert-Laszlo Barabasi
Jorg Menche
Diagnostic Software