This dataset contains results presented in Nawaz, Urbanska, Herbig et al
manuscript entitled ‘ Using real-time fluorescence and deformability
cytometry and deep learning to transfer molecular specificity to label-free
sorting ’ available on bioRxiv (https://doi.org/10.1101/862227) and
includes:
(i) 29 '*.rtdc' files containing measurements performed with real time
fluorescence and deformability cytometry [1,2]
(ii) 1 '*.fcs' file containing a measurement performed with a standard flow
cytometer (BD LSR II, BD Biosciences)
(iii) 1 '*.xlsx' file containing results from AFM-based indentation
experiments.
Sorting real-time fluorescence and deformability cytometry (soRT-FDC) is a
robust sorting platform, combining image-based morphological cell analysis
with mechanical characterisation and subsequent active sorting by feeding
real-time fluorescence and deformability cytometry (RT-FDC) [2] information to
a down-stream SSAW-based cell sorter [3].
Each filename of this dataset starts with the name of the figure to which the
datafile corresponds. The measured samples are briefly described below. The
'Initial' measurements refer to the samples used for sorting, and the 'Target'
measurements to the samples collected in the target outlet after sorting. For
more detailed description of samples please refer to the manuscript.
The '*.rtdc' files are HDF5 files and can be analysed using a Python library
called dclab [4], or a software called Shape-Out [5]. They can also be opened
using other HDF viewer programs.
[1] Otto et al., "Real-time deformability cytometry: on-the-fly cell
mechanical phenotyping". Nature Methods, 12(3):199–202, 2015.
doi:10.1038/nmeth.3281.
[2] Rosendahl et al., "Real-time fluorescence and deformability cytometry".
Nature Methods, 15(5):355–358, 2018. doi:10.1038/nmeth.4639.
[3] Nawaz et al., “Acoustofluidic Fluorescence Activated Cell Sorter”.
Analitycial Chemistry 87(24): 12051–12058, 2015. doi:
doi.org/10.1021/acs.analchem.5b02398
[4] https://github.com/ZellMechanik-Dresden/dclab
[5] https://github.com/ZellMechanik-Dresden/ShapeOut
Figure1d_01_Beads_FL_Initial.rtdc, Figure1d_02_Beads_FL_Target.rtdc
sample: 1:5 mixture of polyacrylamide microgel beads labeled with
AlexaFluor488 and unlabeled ones, these beads were produced in house; sorted
for fluorescence
sorting gates: FL-1 maximum 1000 – 10000, area ratio: 1.0 - 1.1
Figure1d_03_Beads_Size_Initial.rtdc, Figure1d_04_Beads_Size_Target.rtdc
sample: 1:6 mixture of 13.79 ± 0.59 μm silica beads (SiO2-F-L3519-1;
Microparticles, Germany) and 17.23 ± 0.24 μm poly(methyl methacrylate) beads
(PMMM-F-B1423; Microparticles); sorted for size
sorting gates: size 220 – 300 μm2, def 0.00 – 0.01, are ratio 1.0 – 1.1
Figure1d_05_Beads_DefSize_Initial.rtdc,
Figure1d_06_Beads_DefSize_Target.rtdc
sample: a mixture of two polyacrylamide microgel bead populations of different
stiffness, these beads were produced in house; sorted for deformation and size
sorting gates: size 95 – 105 μm2, def 0.000 – 0.016, are ratio 1.0 – 1.1
Figure2b_Blood_Initial_Inlet.rtdc
sample: RBC-depleted blood sample analysed in the inlet region of the chip
Figure2c_Blood_Initial_Channel.rtdc
sample: RBC-depleted blood sample analysed in the usual ROI at the end of the
constricting channel
Figure2d_Blood_Target_RBCs.rtdc
sample: RBC-depleted blood, sorted RBCs
sorting gates: size 25 – 65 μm2, def 0.16 – 0.40, are ratio 1.0 – 1.1
Figure2e_Blood_Target_ly.rtdc
sample: RBC-depleted blood, sorted lymphocytes
sorting gates: size 25 – 45 μm2, def 0.00 – 0.10, are ratio 1.0 – 1.1
Figure2f_Blood_Target_my.rtdc
sample: RBC-depleted blood, sorted myeloid cells
sorting gates: size 53 – 120 μm2, def 0.00 – 0.15, are ratio 1.0 – 1.1
Figure3_Blood_Initial.rtdc, Figure3_Blood_Target_neuBr.rtdc
sample: RBC-depleted blood, brightness-based neutrophils sorting*
sorting gates: size 56 – 100 μm2, brightness 75 – 78, are ratio 1.0 – 1.1
Figure4_Blood_DNN_Training(1).rtdc, Figure4_Blood_DNN_Training(2).rtdc
sample: RBC-depleted blood, datasets for training of deep neural network
(DNN)*
Figure4_Blood_DNN_Validation.rtdc
sample: RBC-depleted blood, dataset for validation of trained DNN*
Figure4d_Blood_Initial.rtdc, Figure4e_Blood_Target_neuDNN.rtdc
sample: RBC-depleted blood, DNN-based neutrophil sorting*
*for identification of neutrophils (CD66+/CD14−), these samples were stained with APC-conjugated anti-human CD14 (dilution 1:20, #17- 809 0149-42, eBioscience, CA, USA; recorded with FL-3 channel) and PE-conjugated anti-human CD66a/c/e (dilution 1:40, 810 #34303, BioLegend, CA, USA; recorded with FL-2 channel); for cross-talk compensation of the fluorescence signal in channels FL-2 and FL-3 for these samples introduce the following correction factor in the ShapeOut software: spill from channel 2 to 3 = 0.025, spill from channel 3 to 2 = 0.100
SFigure2a_BeadMix_RT-FDC.rtdc, SFigure2b_BeadMix_soRT-FDC.rtdc,
SFigure2c_BeadMix_FC.fcs
sample: a mixture of 4 bead types with different diameter, each made of
different material:
9.78 ± 0.15 μm melamine beads (MF-FluoBlau-L948),
13.79 ± 0.59 μm silica beads (SiO2-F-L3519-1),
15.21 ± 0.31 μm polystyrene beads (PS/Q-F-KM194),
17.23 ± 0.24 μm poly(methyl methacrylate) beads (PMMM-F-B1423),
all purchased from Microparticles, Germany.
SFigure3a_Beads_FLDef_Initial.rtdc, SFigure3a_Beads_FLDef_Target.rtdc
sample: a mixture of fluorescent and non-fluorescent polyacrylamide beads with
different mechanical properties, these beads were produced in house; sorted
for fluorescence and deformation
sorting gates: FL-1 maximum 1000 – 10000, def 0.01 – 0.02, are ratio 1.0 – 1.1
SFigure3b_Beads_BrSize_Initial.rtdc, SFigure3b_Beads_BrSize_Target.rtdc
sample: a mixture of 4 bead types as in SFigure2; sorted for brightness and
size
sorting gates: size 160 – 240 μm2, brightness 75 – 85, are ratio 1.0 – 1.1
SFigure4a_Kc167_HL60_Size_Initial.rtdc,
SFigure4b_Kc167_HL60_Size_Target.rtdc
sample: a mixture of Kc167 Drosophila cells and HL60/S4 human promyelocytic
leukaemia cells, sorted for size
sorting gates: size 25 – 77 μm2, def 0.00 – 0.15, are ratio 1.0 – 1.1
SFigure5a_RBC_Def_Initial.rtdc, SFigure5b_RBC_Def_Target1.rtdc,
SFigure5c_RBC_Def_Target2.rtdc
sample: blood anticoagulated with citrate, RBCs sorted for deformation
sorting gates: def 0.15 – 0.40 (Target1) / def 0.00 – 0.10 (Target2), are
ratio 1.0 – 1.2
SFigure7_AFM_AfterSorting.xlsx
sample: HL60/S4 cells collected after sorting experiment; the initial sample
was prepared as for the soRT-FDC experiment but not loaded onto the chip, the
default sample was collected in the default outlet, i.e., run through the
sorting chip but not exposed to SSAW, and the target sample corresponds to
cells exposed to SSAW; reported values correspond to apparent Young’s modulus
estimated from AFM indentation experiments