![]() Alternately, the QC results can be used to exclude aberrant images from full analysis: if using CellProfiler for post-QC data analysis, for example, the classifier produced by the above steps may be incorporated into the CellProfiler analysis pipeline with theįlagImage module. In the absence of such egregious problems, it is helpful to store the QC results as metadata alongside subsequent analysis results to allow for retrospective quality checks and to assist troubleshooting. A systematic microscopy error, for example, could be detected at this point and further downstream data processing could be aborted without further investment of valuable time. ![]() ![]() Our laboratory typically runs the QC workflow prior to completing a full analysis of images using CellProfiler. The QC workflow described above is intended to form the initial steps of a larger data analysis workflow ( see Ref ( 7, 10)). An overview of the workflow is shown in Fig. The results can either be written to a database for further review, or the classifier can be used to filter images within a later CellProfiler pipeline so that only those images which pass QC are used for cellular feature extraction. These measurements are used within the machine-learning tool packaged with CellProfiler Analyst to automatically classify images as passing or failing QC criteria determined by a classification algorithm. The pipeline is then run on the images collected in the experiment to assemble a suite of QC measurements, including the image’s power log-log slope, textural correlation, percentage of the image occupied by saturated pixels, and the standard deviation of the pixel intensities, among others. Image processing modules are selected and placed in the pipeline and the modules’ settings are adjusted appropriately according to the specifics of the HCS project (for example, spatial scales for blur measurements, and the channels used for thresholding see the section “Configuring the The protocol begins with configuring the input and output file locations for the CellProfiler program and constructing a modular QC “pipeline”. The workflow described below expands our prior work using CellProfiler and CellProfiler Analyst validating image-based metrics for QC ( see Ref ( 7)) and provides a step-by-step protocol that leverages the functionality of both of these packages for QC purposes. Likewise, CellProfiler Analyst has been previously used for per-cell classification of phenotypes ( see Ref ( 4, 6)). CellProfiler has been validated for a diverse array of biological applications, typically for generating features on a per-cell basis ( see Ref ( 4, 5)). The protocol uses the open-source, freely downloadable software packages, CellProfiler and CellProfiler Analyst. This chapter outlines a protocol for the characterization of images for common artifacts that confound high-content imaging experiments, including focus blur and image saturation ( Fig. For high-throughput assays, manual inspection of all images for quality control (QC) purposes is not tractable therefore, the development of QC methodologies must be similarly automated to keep up with the increasing demands modern imaging experiments. In our experience, as many as 5% of the fields of view in a routine screen can be affected with such artifacts to varying degrees. ![]() Abnormalities in image quality can degrade otherwise high-quality microscopy data and, in severe cases, even render some experimental approaches infeasible. However, reliable downstream processing of such datasets often depends on robust exclusion of images that would otherwise be erroneously scored as screening hits or inadvertently ignored as false negatives. Analyzing experiments that are comprised of tens to millions of images allows for quantitative modeling of biological processes and discerning complex and subtle phenotypes. Any number of high-content assays can be quantified by combining high-resolution microscopy with sophisticated image analysis techniques in order to create an automated workflow with a high degree of reproducibility, fidelity, and robustness ( see Ref ( 3)). The use of automated microscopy combined with image analysis methods has enabled the extraction of quantitative image-based information from cells, tissues, and organisms while speeding analysis and reducing subjectivity ( see Ref ( 1, 2)). ![]()
0 Comments
Leave a Reply. |