Cytokit: A single-cell analysis toolkit for high dimensional fluorescent microscopy imaging

Abstract

BackgroundMultiplexed in-situ fluorescent imaging offers several advantages over single-cell assays that do not preserve the spatial characteristics of biological samples. This spatial information, in addition to morphological properties and extensive intracellular or surface marker profiling, comprise promising avenues for rapid advancements in the understanding of disease progression and diagnosis. As protocols for conducting such imaging experiments continue to improve, it is the intent of this study to provide and validate software for processing the large quantity of associated data in kind. ResultsCytokit offers (i) an end-to-end, GPU-accelerated image processing pipeline; (ii) efficient input/output (I/O) strategies for operations specific to high dimensional microscopy; and (iii) an interactive user interface for cross filtering of spatial, graphical, expression, and morphological cell properties within the 100+ GB image datasets common to multiplexed immunofluorescence.Image processing operations supported in Cytokit are generally sourced from existing deep learning models or are at least in part adapted from open source packages to run in a single or multi-GPU environment. The efficacy of these operations is demonstrated through several imaging experiments that pair Cytokit results with those from an independent but comparable assay. A further validation also demonstrates that previously published results can be reproduced from a publicly available multiplexed image dataset.

Mammalian gene expression variability is explained by underlying cell state

“…The high goodness of fit of the two-state model to uncorrected or partially corrected distributions that shows substantial bursting could simply be a case of over interpretation of model fit. RNA binding systems, such as MS2, allow direct live-cell observation of transcription bursting, and many groups have observed burst-like punctuated transcription (Corrigan et al, 2016;Ferguson and Larson, 2013;Fritzsch et al, 2018;Muramoto et al, 2010) . While direct visualization is compelling, it is unclear if punctuated transcriptional events are due to stochastic transition of promoter state, as suggested by two state model, or due to stochasticity in the activity of an upstream regulatory element.…”

“…One interpretation is that the observed over-dispersion is simply a result of the superposition of an allele-specific Poisson variability and cell state variability (Battich et al, 2015) . The other interpretation is that the allele-specific variability itself is not a simple Poisson process (Corrigan et al, 2016;Dar et al, 2015;Suter et al, 2011;Tantale et al, 2016) . The latter interpretation was popularized by the introduction of a simple phenomenological model named the two-state or random telegraph model that represented genes as existing in either "on" or "off" states (Friedman et al, 2006;Fukaya et al, 2016;Kaern et al, 2005;Kepler and Elston, 2001;Lenstra et al, 2016;Molina et al, 2013;Paulsson, 2004;Peccoud and Ycart, 1995;Raj et al, 2006;Sanchez and Golding, 2013;Suter et al, 2011;Thattai and van Oudenaarden, 2004) .…”

“…The latter interpretation was popularized by the introduction of a simple phenomenological model named the two-state or random telegraph model that represented genes as existing in either "on" or "off" states (Friedman et al, 2006;Fukaya et al, 2016;Kaern et al, 2005;Kepler and Elston, 2001;Lenstra et al, 2016;Molina et al, 2013;Paulsson, 2004;Peccoud and Ycart, 1995;Raj et al, 2006;Sanchez and Golding, 2013;Suter et al, 2011;Thattai and van Oudenaarden, 2004) . More complex models with multiple states were also considered, (Corrigan et al, 2016;Nicolas et al, 2018;Suter et al, 2011;Tantale et al, 2016;Zoller et al, 2015) but the addition of multiple states does not change the model in a qualitative way. These models suggest that transcription should occur in distinct bursts with multiple transcripts generated when the gene is "on".…”

Towards a scaled-up T cell-mediated cytotoxicity assay in 3D cell culture using microscopy

“…The two other methods employed extend this process by attempting to further identify individual cells within spheroids. The first of these uses the U-Net-based nuclei segmentation model of CellProfiler (Caicedo et al 2019) via Cytokit (Czech et al 2018) while the second utilizes a custom cytometry function based on differences of gaussian filters. This second method affords greater flexibility in parameterization to aid in tuning thresholds (albeit subjectively) to better capture nuclei images with noisy, occluded boundaries.…”

Viable and efficient electroporation-based genetic manipulation of unstimulated human T cells

“…See this notebook for detailed data and analysis. (b) Cytokit was used to analyze the microscope images (Czech et al, 2018) . Each sub-panel shows 60 representative individual cells that were plasmid positive.…”