BI-SCUDO (biomarker identification-sample classification SCUDO) is a normalization-free, rank-based tool to classify expression profiles, typically for diagnostic/prognostic purposes. Given a set of expression profiles, one per subject, the tool works by first extracting a rank-based signature given by the identifiers of the n1 most expressed probes and the n2 least expressed probes. Subject signatures are then compared by means of a distance measure based on the definition of weighted enrichment score to obtain a signature distance matrix, which is then used to infer a classification of the profiles according to a phenotype of interest (such as control/disease status). Classification results are then displayed by a new graphical formalism that produces an unambiguous and easy to read sample-to-group assignment chart. Statistics and p-values are automatically computed to assess the quality of the output and to enable comparisons with other methods. Results include also a biomarker panel which clearly identifies the list of gene/miRNA/protein levels that BI-SCUDO uses to compute sample classification.
HOW TO CITE AND REFERENCES
BI-SCUDO, COSBI, The Microsoft Research - University of Trento Centre for Computational and Systems Biology, http://www.cosbi.eu/research/prototypes/BI-SCUDO
Lauria M. Rank-based transcriptional signatures: a novel approach to diagnostic biomarker definition and analysis, Systems Biomedicine 1(4):35-46, 2013.
Lauria M., Moyseos P., Priami C. SCUDO: a tool for signature-based clustering of expression profiles, Nucleic acids research 43 (W1), W188-W192, 2015.
Caberlotto L. et al. Integration of transcriptomic and genomic data suggests candidate mechanisms for APOE4-mediated pathogenic action in Alzheimer's disease, Scientific Reports 6, 2016.
Lacroix S. et al. Systems biology approaches to study the molecular effects of caloric restriction and polyphenols on aging processes, Genes & Nutrition 10(6):1-10, 2016.
Please address all enquiries about BI-SCUDO to the BI-SCUDO Team