Screening cellular feature measurements for image-based assay development

Abstract

The typical "design" approach to image-based assay development involves choosing measurements that are likely to correlate with the phenotype of interest, based on the researcher's intuition and knowledge of image analysis. An alternate "screening" approach is to measure a large number of cellular features and systematically test each feature to identify those that are best able to distinguish positive and negative controls while taking precautions to avoid overfitting the available data. The cell measurement software the authors previously developed, CellProfiler, makes both approaches straightforward, easing the process of assay development. Here, they demonstrate the use of the screening approach to image assay development to select the best measures for scoring publicly available image sets of 2 cytoplasm-to-nucleus translocation assays and 2 Transfluor assays. The authors present the resulting assay quality measures as a baseline for future algorithm comparisons, and all software, methods, and images they present are freely available.

Fig. 1.

Image assays and image analysis pipeline design. (A) Two representative images from each of the 4 image sets are shown: 1 positive and 1 negative control. Each has a red or blue channel (DNA stain) and a green channel (labeling the molecule of interest). (B) A schematic pipeline, displaying typical steps of the image analysis pipeline.

Fig. 2.

Cytoplasm-to-nucleus translocation (CNT) assay quality statistics. (A, left) Bar plots display Z′ factors for each feature, grouped by feature category and ordered by descending Z′ factor values within each category. Values below –10 are not displayed, and note that the maximum possible Z′ factor is 1. Feature categories begin at their x-axis label and progress to the right. BioImage LY294002 and Vitra MCF7 plots are not shown, as they are qualitatively similar to the respective plots shown. (A, right) The same data are plotted, but values below 0 are not shown, highlighting features that are potentially screenable. The 2 CNT data sets display similar screenable features, and V factors also display qualitatively the same pattern (not shown). Note that the number of intensity measures are greater in BioImage than in Vitra because the propagate algorithm was not reliable at detecting the cytoplasm in Vitra positive control images and thus not screened. (B) The top-scoring feature, in terms of Z′ and V factor, is shown for each CNT image set. BioImage has 2 drugs, wortmannin and LY294002, and Vitra has 2 cell types, A549 and MCF7.

Fig. 3.

Transfluor assay quality statistics. Layout is the same as Figure 2. Note that values below –10 are not displayed, and this varied between the 2 data sets, explaining the seeming difference in numbers of features within each category.