Kurlovs, A. | bioRxiv Science

bioRxiv2025

Single-cell neural network classifiers reveal that PM21 NK cell expansion is dependent on B cell signaling

In the field drug development ML/AI methods are being applied to improve drug production speed, costs, and reliability. In allogenic NK cell therapy production, one of the biggest challenges is the inherent variability in the donors that provide the starting material for NK cell expansion. In this study we performed PM21-mediated NK cell expansion on 26 donors, and in parallel performed single-cell transcriptomics on the same donor sample prior to expansion. Canonical differential expression analysis and cell state abundance did not highlight any significant difference between donors with high and low NK cell expansion yield. Instead, training neural networks classifiers for high-yield donors enabled identifying several highly predictive models with perfect cross-validation recall. Further investigation of the most predictive models unveiled a previously unknown role for B cell in supportive NK cell expansion. Overall, this study represents a blueprint for combining deep phenotyping and machine learning methods to unveil novel biology and improve the quality and speed of delivery of cell therapeutics to patients.

systems biology↗

bioRxiv2022

DiSiR: a software framework to identify ligand-receptor interactions at subunit level from single-cell RNA-sequencing data

Most of cell-cell interactions and crosstalks are mediated by ligand-receptor interactions. The advent of single-cell RNA-sequencing (scRNA-seq) techniques has enabled characterizing tissue heterogeneity at single-cell level. Over the past recent years, several methods have been developed to study ligand-receptor interactions at cell type level using scRNA-seq data. However, there is still no easy way to query the activity of a specific user-defined signaling pathway in a targeted way or map the interactions of the same subunit with different ligands as part of different receptor complexes. Here, we present DiSiR, a fast and easy-to-use permutation-based software framework to investigate how individual cells are interacting with each other by analyzing signaling pathways of multi-subunit ligand-activated receptors from scRNA-seq data, not only for available curated databases of ligand-receptor interactions, but also for interactions that are not listed in these databases. We show that, when utilized to infer melanoma disease map on a gold-standard dataset, DiSiR outperforms other well-known permutation-based methods, e.g., CellPhoneDB and ICELLNET. To demonstrate DiSiRs utility in exploring data and generating biologically relevant hypotheses, we apply it to COVID lung and rheumatoid arthritis (RA) synovium scRNA-seq data and highlight potential differences between inflammatory pathways at cell type level for control vs. disease samples.

cell biology↗