CellProfiler Analyst: interactive data exploration, analysis, and classification of large biological image sets

Abstract

Summary: CellProfiler Analyst allows the exploration and visualization of image-based data, together with the classification of complex biological phenotypes, via an interactive user interface designed for biologists and data scientists. CellProfiler Analyst 2.0, completely rewritten in Python, builds on these features and adds enhanced supervised machine learning capabilities (Classifier), as well as visualization tools to overview an experiment (Plate Viewer and Image Gallery). Availability and Implementation: CellProfiler Analyst 2.0 is free and open source, available at

In vivo identification of apoptotic and extracellular vesicle‐bound live cells using image‐based deep learning

“…To compare to classical machine learning, the Cell‐Profiler (CP) [29] pipeline from Blasi et al [28] was used for feature extraction. However, in our case the second channel corresponds to fluorescence intensity instead of darkfield.…”

“…Upon intravenous injection of MFG‐E8‐eGFP we performed imaging flow cytometry of fresh tissue cells on an ImageStream x MarkII imaging cytometer, which allows detection of small particles with high sensitivity [23] and generates detailed images of individual cells [24]. To automatically classify apoptotic vs. EV‐decorated (EV + ) cells, we developed a convolutional autoencoder (CAE) [25–27], which combines the advantages of traditional feature extraction [28–30] and deep learning [31] for imaging flow cytometry. Using this pipeline, we show that MFG‐E8‐eGFP detects apoptotic as well as EV + cells in vivo .…”

Melanoma Persister Cells Are Tolerant to BRAF/MEK Inhibitors via ACOX1-Mediated Fatty Acid Oxidation

“…FFPE sections were stained with an antibody recognizing PPARa using an automated image analysis with CellProfiler (Carpenter et al, 2006) (Figure S7A). To quantify the total numbers of PPARa + cells and PPARa À cells in each tumor biopsy, we used CellAnalyst software, which is designed to explore image-based data (Dao et al, 2016), to train a set of manually selected PPARa-positive or PPARa-negative cells and then classified all the cells from the 14 tumor biopsies as PPARa + or PPARa À ( Figure S7B). We found that either high expression of PPARa at baseline or an increase of PPARa expression 2 weeks after treatment initiation was associated with non-responding tumors, whereas low basal PPARa expression or decreased PPARa expression after treatment were associated with response to targeted therapy samples ( Figures 7D, S7C, and S7D).…”

Toxoplasma gondii infection induces the formation of host’s nuclear granules containing poly(A)-binding proteins

“…The coverslips were mounted onto microscope slides with Fluoromount-G (Electron Microscopy Sciences, 17984-25), and viewed by fluorescence microscopy (Leica DMI 6000 fluorescence microscope with Leica Application Suite Advanced Fluorescence Software). Approximately 100 independent fields of view were taken, and the images were analyzed by CellProfiler 2.2.0 (Dao et al 2016;Jones et al 2008). Data were compiled into Microsoft Excel spreadsheets, then averages and standard errors were calculated for each infection time point.…”