iNoisET

iNoisET implements a generalized Bayesian framework to model biological and experimental noise in high-throughput T- and B-cell receptor repertoire sequencing and to detect statistically significant responding clonotypes.

Key Features:

• Noise Characterization: Implements a Bayesian method to model both biological and experimental noise, accounting for sampling, library preparation, and expression noise in repertoire sequencing data.
• Experimental Noise Model Learning: Learns experimental noise models from replicate datasets to characterize technical variability.
• Detection of Responding Clonotypes: Identifies clonotypes that respond to stimuli such as vaccines or disease conditions by detecting statistically significant changes.
• Versatility Across Sequencing Technologies: Applicable to and tested on multiple repertoire sequencing technologies and diverse datasets.

Scientific Applications:

• Immune repertoire analysis: Quantitative analysis of T- and B-cell receptor repertoires to study immune responses.
• Longitudinal and spatial studies: Investigation of immune dynamics over time and across different tissues.
• Disease and vaccination studies: Analysis of clonal expansions and contractions in acute and chronic diseases and assessment of vaccine-induced responses.

Methodology:

Uses a generalized Bayesian framework that models biological and experimental noise, learns noise models from replicate datasets, integrates these noise models into the analysis, and detects significant clonotype changes following stimuli.