GLaMST

GLaMST reconstructs lineage trees from B cell receptor (BCR) sequencing data to infer affinity maturation and microevolutionary dynamics of immunoglobulin genes.

Key Features:

• Lineage tree reconstruction: Reconstructs lineage trees from observed BCR sequences where nodes represent specific BCR sequences and directed edges indicate single base substitutions, insertions, or deletions, accounting for partially observed nodes in the microevolutionary process.
• Algorithmic performance: Demonstrated improved performance on simulated and real BCR data compared to existing algorithms, particularly under high rates of mutation, insertion, and deletion, yielding lineage trees that are smaller and closer to ground truth.
• Correlation with selection pressure: Produces lineage-tree features that correlate with selection pressure, facilitating analysis of evolutionary dynamics acting on immunoglobulin genes.
• Interoperability: Designed to integrate with existing BCR sequencing analysis frameworks for use within broader analytical pipelines.

Scientific Applications:

• Affinity maturation analysis: Reconstruction of B cell lineage trees to study somatic hypermutation, selection, and the evolutionary pathways that produce high-affinity antibodies.

Methodology:

GLaMST grows lineages along a minimum spanning tree derived from observed BCR sequences, treating observed sequences as partial nodes and inferring directed edges that represent single nucleotide substitutions, insertions, and deletions.