CytoBatchFlagR: Spectral flow cytometry dataset

Recent advancements in flow and mass cytometry techniques have led to the generation of high-parameter (>40 markers) single-cell datasets, often used to create large scale, longitudinal studies with possible technical variabilities occurring between experimental runs. Many state of the art computational methods to correct for these batch effects exist but do not provide an automated assessment of batch effects before correction. Here, we developed CytoBatchFlagR, an automated computation framework that highlights problematic batches and markers within a dataset, guiding researchers to objectively inspect their data before the exclusion of these batches/markers or applying a batch normalization algorithm.

Technical controls were obtained from healthy, non-pregnant adult peripheral blood donors at a local blood donation center. Peripheral leukocytes were isolated by Ficoll gradient centrifugation for 20 min at 800g). Leukocytes from each donor were washed once with wash medium (RPMI 10% FBS pen/strep) and centrifugated for 7min at 650g). Unstimulated controls tcu1 and tcu2 were directly aliquoted in X-VIVO supplemented with 10% FBS and 10% DMSO and frozen in liquid nitrogen. Stimulated controls tcs3 and tcs4  were cultured in X-VIVO medium supplemented with 5% human serum and 50 U/mL IL-2 in the presence ofanti-CD3/CD28 beads (BD Biosciences, 1ug/ml) for 7 days. Leukocytes were collected on day 7 post-stimulation, aliquotted and frozen in liquid nitrogen as described. Controls were thawed at 37°C,  washed once and directly stained for analysis on a Cytek Aurora spectral flow cytometer. For surface staining, cells were stained for 30 min in RPMI containing 1% FBS and pen/strep. For intracellular staining, cells were fixed and permeabilized using the eBiosciences FOXP3 kit (eBiosciences). Flow cytometry data was acquired in 18 batches using equipment maintained by the Research Flow Cytometry Core in the Division of Rheumatology at Cincinnati Children’s Hospital Medical Center. Initial data processing and gating on CD45+CD14- live single cells was performed using FlowJo software[version]. FCS files of CD45+CD14- live single cells were exported for further analysis in R software.

Details on antibody clones and concentrations as used for staining are included as Table 1. The meta datatable that includes all information including file name, sample ID, batch number as required for R-based analysis including CytoBatchFlagR are included in Table 2.