SCUDO

SCUDO performs signature-based clustering of gene expression profiles to classify and diagnose individuals by comparing subject-specific mRNA and miRNA transcriptional signatures against a reference map.

Key Features:

• Subject-specific transcriptional signatures: Uses subject-specific rather than disease-specific transcriptional signatures for analysis.
• Reference transcriptional map: Constructs a reference map composed of transcriptional signatures derived from mRNA and miRNA profiles of healthy and affected individuals.
• Rank-based signatures: Employs rank-based signatures to enhance detection of subtle variations in gene expression profiles.
• Positional classification: Determines the position of an individual's transcriptional signature within the reference map for clustering-based classification and diagnosis.
• Signature-based clustering: Performs clustering of gene expression profiles using signature comparisons.
• Empirical validation: Validated performance in the SBV Improver Diagnostic Signature Challenge, achieving second place overall and first place in a sub-challenge.

Scientific Applications:

• Diagnostic classification: Classifies and aids diagnosis of individuals using mRNA and miRNA transcriptional profiles.
• Precision medicine: Supports development of individualized transcriptional biomarkers for precision medicine applications.
• Gene expression analysis benchmarking: Provides a framework for benchmarking signature-based methods in challenges such as the SBV Improver Diagnostic Signature Challenge.

Methodology:

Construction of a reference map from transcriptional signatures derived from mRNA or miRNA profiles, generation of rank-based subject-specific signatures, and positional comparison of a query individual's signature within the reference map for clustering-based classification.