MathIOmica

MathIOmica analyzes longitudinal omics — including transcriptomics, proteomics, and metabolomics — and generalized time series data to classify temporal patterns and assess biological significance.

Key Features:

• Supported data types: Handles longitudinal transcriptomics, proteomics, metabolomics and generalized time series datasets.
• Data import and preprocessing: Provides import utilities along with quality control and normalization for prepared downstream analysis.
• Time series generation and classification: Generates time series from input data and classifies temporal trends using spectral methods such as periodograms and autocorrelations.
• Handling missing and uneven data: Includes methods to address missing data points and uneven sampling intervals in longitudinal studies.
• Visualization: Produces collective temporal visualizations such as heatmaps and other graphical representations for interpretation of temporal patterns.
• Biological significance assessment: Performs Gene Ontology (GO) and pathway enrichment analyses to link temporal trends to biological processes and pathways.
• Statistical validation: Generates null distributions using randomly resampled time series to establish statistical significance cutoffs for classifications.

Scientific Applications:

• Individualized profiling: Identifies significant temporal trends in longitudinal experimental datasets for subject-level molecular profiling.
• Personalized health monitoring: Detects temporal patterns potentially associated with adverse health events or other clinically relevant outcomes in longitudinal monitoring.

Methodology:

Implements data import, quality control, normalization, time-series generation, spectral analyses (periodograms and autocorrelations), classification of temporal trends, handling of missing/uneven sampling, visualization (heatmaps), Gene Ontology and pathway enrichment, and null-distribution generation via random resampling.