A1138
Title: FiXeD: Spatial point process distances for pairing the heavy and light chains of B cell receptors from spatial BCR-seq
Authors: Yimeng Liu - University of Victoria (Canada) [presenting]
Abstract: The immune system identifies and mounts a defense against tumor cells through antigen recognition, a process mediated primarily by T cells and B cells. B cells are equipped with heavy and light receptor chains, whereas T cells carry alpha and beta receptor chains. These receptors recognize antigens and orchestrate an immune response, making their accurate pairing essential for understanding tumor-immune interactions. The purpose is to develop an approach based on spatial transcriptomic data and BCR-seq to infer receptor chain pairings. The approach is formulated as a combinatorial optimization problem with an objective function that incorporates the expression matrices of heavy and light chains with their spatial co-expression point patterns derived from barcoded pixels. Spatial information is represented as distances between point patterns computed using an optimal transport algorithm. The estimated receptor heavy and light chain pairs have potential application in the development of targeted immunotherapies. The methodology is evaluated using both simulation studies based on ovarian and breast cancer cohorts, as well as with real data, where the ground truth is known from single-cell sequencing. Substantial improvements in terms of both accuracy and stability are demonstrated over existing state-of-the-art.
