Technical Overview
This section presents the architectural and algorithmic foundation of AlignAIR. It is intended for users who are interested in understanding the inner workings of the pipeline including its segmentation model, likelihood thresholding mechanism, and mutation simulation framework.
🧠 Model Architecture
Learn how AlignAIR's residual convolutional backbone processes and segments sequences.
Explore Architecture →🎚️ Thresholding Logic
Understand how AlignAIR selects V, D, and J alleles from likelihood vectors using dynamic thresholds and caps.
Explore Thresholding →🧬 Mutation Models
Explore the simulation models used to train AlignAIR on realistic SHM patterns, including S5F.
Explore Mutation Models →