Technical Overview

Architectural and algorithmic foundation of AlignAIR. Intended for users who want to understand the segmentation model, likelihood thresholding mechanism, and mutation simulation framework.

01 / architecture

Model architecture

Learn how AlignAIR's residual convolutional backbone processes and segments sequences.

Open

02 / thresholding

Thresholding logic

How AlignAIR selects V, D, and J alleles from likelihood vectors using dynamic thresholds and caps.

Open

03 / mutations

Mutation models

Simulation models used to train AlignAIR on realistic SHM patterns, including S5F.

Terms·License·Open Source