Increasing the Content of High-Content Screening: An Overview

Abstract

Target-based high-throughput screening (HTS) has recently been critiqued for its relatively poor yield compared to phenotypic screening approaches. One type of phenotypic screening, image-based high-content screening (HCS), has been seen as particularly promising. In this article, we assess whether HCS is as high content as it can be. We analyze HCS publications and find that although the number of HCS experiments published each year continues to grow steadily, the information content lags behind. We find that a majority of high-content screens published so far (60-80%) made use of only one or two image-based features measured from each sample and disregarded the distribution of those features among each cell population. We discuss several potential explanations, focusing on the hypothesis that data analysis traditions are to blame. This includes practical problems related to managing large and multidimensional HCS data sets as well as the adoption of assay quality statistics from HTS to HCS. Both may have led to the simplification or systematic rejection of assays carrying complex and valuable phenotypic information. We predict that advanced data analysis methods that enable full multiparametric data to be harvested for entire cell populations will enable HCS to finally reach its potential.

Keywords: Cell-based assays; high-content screening; image analysis; phenotypic drug discovery; statistical analyses.

Figure 1.

The number of papers in which a high-throughput, image-based experiment was used toward a discovery, by year of publication. Combined indicates the sum of all three searches. Note that the Combined trend line should not be considered as a total, because the literature searches are not at all comprehensive.

Figure 2.

Feature set sizes used in papers throughout the three searches. Numbers at the top of each bar indicate the actual number of papers. Between 60% and 80% of the papers used only one or two measured features of the cells.

Figure 3.

Percentage of papers that use only one or two measured features of the cells, by year of publication.

Figure 4.

The percentage of papers throughout all three searches that use the Z′-factor, plotted by year of publication. The fractions indicate the number of papers that use the Z′-factor divided by the total number of papers in each year. Overall, 40% use the Z′-factor (dotted line).

Figure 5.

The necessary steps required to use the Z′-factor as a quality metric drastically simplify assay readout and analysis but typically also reduce the power and value of an HCS assay. (A) The Z′-factor is a univariate statistic, so assay developers typically select a single feature as a readout, ignoring a large part of other available information; (B) the per-cell measurements need to be aggregated into a single value per replicate sample, and assays presenting heterogeneous cell responses detectable only via subtleties in their population distributions will often fail to yield acceptable Z′-factor values and be discarded; and (C) the Z′-factor requires that the distributions of controls’ values are Gaussian—a condition that is met by choosing the method of aggregation to be the mean throughout the cell population—but this biases the selection of assays considerably, as discussed in the text.