iMEGES

iMEGES prioritizes susceptibility genes and genetic variants associated with mental disorders by integrating coding and non-coding variants, structural variants (SVs), brain expression quantitative trait loci (eQTLs), and epigenetic data using deep neural networks implemented in TensorFlow.

Key Features:

• Deep Neural Network Integration: Utilizes deep neural networks implemented in TensorFlow to analyze complex genomic and epigenomic datasets.
• Comprehensive Data Input: Accepts genetic mutations and phenotypic information and integrates coding and non-coding variants as well as structural variants (SVs).
• Variant and Gene Prioritization: Produces ranked lists of susceptibility variants and prioritized disease-specific genes by integrating brain expression quantitative trait loci (eQTLs) and epigenetic information from PsychENCODE.
• Enhanced Analytical Capability: Demonstrates improved performance over existing methods when trained on large datasets, supporting both population-level and individual patient analyses.

Scientific Applications:

• Population Studies: Aids identification of novel genes or variants that contribute to mental disorder susceptibility across diverse populations.
• Personalized Medicine: Identifies relevant genetic factors in individual patients to inform tailored therapeutic strategies.

Methodology:

Deep neural networks implemented in TensorFlow; integration of coding and non-coding variants, structural variants (SVs), brain eQTLs, and epigenetic data from PsychENCODE; training on large datasets; outputting ranked lists of susceptibility variants and prioritized disease-specific genes.