
Kelvin C. Mo
Computational immunology researcher focused on CAR T-cell therapy
Summary
Work
Education
Projects
Writing
Patient-level prediction from single-cell data using attention-based multiple instance learning with regulatory priors (tcellMIL)
January 1, 2025Describes tcellMIL, an attention-based multiple instance learning framework using regulon activity priors to predict patient-level outcomes from single-cell data and identify outcome-relevant subpopulations; applied to CAR T infusion products.
CCL8/CCL13-Producing Tumor-Associated Macrophages Linked to Poor Outcomes after CAR T Cell Therapy for LBCL
November 1, 2024Oral abstract reporting multi-omic single-cell analyses that implicate CCL8/CCL13-producing tumor-associated macrophages in poor outcomes after CAR T cell therapy for large B-cell lymphoma (LBCL); includes single-cell interactome and deconvolution analyses validating associations with treatment resistance.
CD22-Directed CAR T Cell Single Cell Multiomic Features Associated with Immune Effector Cell-Associated Hemophagocytic Lymphohistiocytosis-like Syndrome (IEC-HS)
November 1, 2024Multiomic single-cell study identifying transcriptional and epigenetic features in CAR22 products and expanding CAR22 cells associated with IEC-HS, highlighting interferon-stimulated genes, STAT2 activity, and chromatin accessibility signatures.