Robust hit identification by quality assurance and multivariate data analysis of a high-content, cell-based assay

Recent technological advances in high-content screening instrumentation have increased its ease of use and throughput, expanding the application of high-content screening to the early stages of drug discovery. However, high-content screens produce complex data sets, presenting a challenge for both extraction and interpretation of meaningful information. This shifts the high-content screening process bottleneck from the experimental to the analytical stage. In this article, the authors discuss different approaches of data analysis, using a phenotypic neurite outgrowth screen as an example. Distance measurements and hierarchical clustering methods lead to a profound understanding of different high-content screening readouts. In addition, the authors introduce a hit selection procedure based on machine learning methods and demonstrate that this method increases the hit verification rate significantly (up to a factor of 5), compared to conventional hit selection based on single readouts only.     

Cellular phenotype recognition for high-content RNA interference genome-wide screening

Genome-wide, cell-based screens using high-content screening (HCS) techniques and automated fluorescence microscopy generate thousands of high-content images that contain an enormous wealth of cell biological information. Such screens are key to the analysis of basic cell biological principles, such as control of cell cycle and cell morphology. However, these screens will ultimately only shed light on human disease mechanisms and potential cures if the analysis can keep up with the generation of data. A fundamental step toward automated analysis of high-content screening is to construct a robust platform for automatic cellular phenotype identification. The authors present a framework, consisting of microscopic image segmentation and analysis components, for automatic recognition of cellular phenotypes in the context of the Rho family of small GTPases. To implicate genes involved in Rac signaling, RNA interference (RNAi) was used to perturb gene functions, and the corresponding cellular phenotypes were analyzed for changes. The data used in the experiments are high-content, 3-channel, fluorescence microscopy images of Drosophila Kc167 cultured cells stained with markers that allow visualization of DNA, polymerized actin filaments, and the constitutively activated Rho protein Rac(V12). The performance of this approach was tested using a cellular database that contained more than 1000 samples of 3 predefined cellular phenotypes, and the generalization error was estimated using a cross-validation technique. Moreover, the authors applied this approach to analyze the whole high-content fluorescence images of Drosophila cells for further HCS-based gene function analysis.     

The FLIGHT Drosophila RNAi database

FLIGHT (http://flight.icr.ac.uk/) is an online resource compiling data from high-throughput Drosophila in vivo and in vitro RNAi screens. FLIGHT includes details of RNAi reagents and their predicted off-target effects, alongside RNAi screen hits, scores and phenotypes, including images from high-content screens. The latest release of FLIGHT is designed to enable users to upload, analyze, integrate and share their own RNAi screens. Users can perform multiple normalizations, view quality control plots, detect and assign screen hits and compare hits from multiple screens using a variety of methods including hierarchical clustering. FLIGHT integrates RNAi screen data with microarray gene expression as well as genomic annotations and genetic/physical interaction datasets to provide a single interface for RNAi screen analysis and data-mining in Drosophila.     

High-throughput RNAi screening by time-lapse imaging of live human cells

RNA interference (RNAi) is a powerful tool to study gene function in cultured cells. Transfected cell microarrays in principle allow high-throughput phenotypic analysis after gene knockdown by microscopy. But bottlenecks in imaging and data analysis have limited such high-content screens to endpoint assays in fixed cells and determination of global parameters such as viability. Here we have overcome these limitations and developed an automated platform for high-content RNAi screening by time-lapse fluorescence microscopy of live HeLa cells expressing histone-GFP to report on chromosome segregation and structure. We automated all steps, including printing transfection-ready small interfering RNA (siRNA) microarrays, fluorescence imaging and computational phenotyping of digital images, in a high-throughput workflow. We validated this method in a pilot screen assaying cell division and delivered a sensitive, time-resolved phenoprint for each of the 49 endogenous genes we suppressed. This modular platform is scalable and makes the power of time-lapse microscopy available for genome-wide RNAi screens.     

The FLIGHT Drosophila RNAi database: 2010 update

FLIGHT (http://flight.icr.ac.uk/) is an online resource compiling data from high-throughput Drosophila in vivo and in vitro RNAi screens. FLIGHT includes details of RNAi reagents and their predicted off-target effects, alongside RNAi screen hits, scores and phenotypes, including images from high-content screens. The latest release of FLIGHT is designed to enable users to upload, analyze, integrate and share their own RNAi screens. Users can perform multiple normalizations, view quality control plots, detect and assign screen hits and compare hits from multiple screens using a variety of methods including hierarchical clustering. FLIGHT integrates RNAi screen data with microarray gene expression as well as genomic annotations and genetic/physical interaction datasets to provide a single interface for RNAi screen analysis and datamining in Drosophila.Key words: RNAi, database, integration, bioinformatics, phenotype     

A picture is worth a thousand words: Genomics to phenomics in the yeast Saccharomyces cerevisiae

Large scale cell biological experiments are beginning to be applied as a systems-level approach to decipher mechanisms that govern cellular function in health and disease. The use of automated microscopes combined with digital imaging, machine learning and other analytical tools has enabled high-content screening (HCS) in a variety of experimental systems. Successful HCS screens demand careful attention to assay development, data acquisition methods and available genomic tools. In this minireview, we highlight developments in this field pertaining to yeast cell biology and discuss how we have combined HCS with methods for automated yeast genetics (synthetic genetic array (SGA) analysis) to enable systematic analysis of cell biological phenotypes in a variety of genetic backgrounds.     

A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen

To enable arrayed or pooled loss-of-function screens in a wide range of mammalian cell types, including primary and nondividing cells, we are developing lentiviral short hairpin RNA (shRNA) libraries targeting the human and murine genomes. The libraries currently contain 104,000 vectors, targeting each of 22,000 human and mouse genes with multiple sequence-verified constructs. To test the utility of the library for arrayed screens, we developed a screen based on high-content imaging to identify genes required for mitotic progression in human cancer cells and applied it to an arrayed set of 5,000 unique shRNA-expressing lentiviruses that target 1,028 human genes. The screen identified several known and approximately 100 candidate regulators of mitotic progression and proliferation; the availability of multiple shRNAs targeting the same gene facilitated functional validation of putative hits. This work provides a widely applicable resource for loss-of-function screens, as well as a roadmap for its application to biological discovery.     

"Plate cherry picking": a novel semi-sequential screening paradigm for cheaper, faster, information-rich compound selection

This work describes a novel semi-sequential technique for in silico enhancement of high-throughput screening (HTS) experiments now employed at Novartis. It is used in situations in which the size of the screen is limited by the readout (e.g., high-content screens) or the amount of reagents or tools (proteins or cells) available. By performing computational chemical diversity selection on a per plate basis (instead of a per compound basis), 25% of the 1,000,000-compound screening was optimized for general initial HTS. Statistical models are then generated from target-specific primary results (percentage inhibition data) to drive the cherry picking and testing from the entire collection. Using retrospective analysis of 11 HTS campaigns, the authors show that this method would have captured on average two thirds of the active compounds (IC(50) < 10 microM) and three fourths of the active Murcko scaffolds while decreasing screening expenditure by nearly 75%. This result is true for a wide variety of targets, including G-protein-coupled receptors, chemokine receptors, kinases, metalloproteinases, pathway screens, and protein-protein interactions. Unlike time-consuming "classic" sequential approaches that require multiple iterations of cherry picking, testing, and building statistical models, here individual compounds are cherry picked just once, based directly on primary screening data. Strikingly, the authors demonstrate that models built from primary data are as robust as models built from IC(50) data. This is true for all HTS campaigns analyzed, which represent a wide variety of target classes and assay types.     

Systematic epistatic mapping of cellular processes

Genetic screens have identified many novel components of various biological processes, such as components required for cell cycle and cell division. While forward genetic screens typically generate unstructured ‘hit’ lists, genetic interaction mapping approaches can identify functional relations in a systematic fashion. Here, we discuss a recent study by our group demonstrating a two-step approach to first screen for regulators of the mitotic cell cycle, and subsequently guide hypothesis generation by using genetic interaction analysis. The screen used a high-content microscopy assay and automated image analysis to capture defects during mitotic progression and cytokinesis. Genetic interaction networks derived from process-specific features generate a snapshot of functional gene relations in those processes, which follow a temporal order during the cell cycle. This complements a recently published approach, which inferred directional genetic interactions reconstructing hierarchical relationships between genes across different phases during mitotic progression. In conclusion, this strategy leverages unbiased, genome-wide, yet highly sensitive and process-focused functional screening in cells.     

The complete automation of cell culture: improvements for high-throughput and high-content screening

Genomic approaches provide enormous amounts of raw data with regard to genetic variation, the diversity of RNA species, and protein complement. High-throughput (HT) and high-content (HC) cellular screens are ideally suited to contextualize the information gathered from other "omic" approaches into networks and can be used for the identification of therapeutic targets. Current methods used for HT-HC screens are laborious, time-consuming, and prone to human error. The authors thus developed an automated high-throughput system with an integrated fluorescent imager for HC screens called the AI.CELLHOST. The implementation of user-defined culturing and assay plate setup parameters allows parallel operation of multiple screens in diverse mammalian cell types. The authors demonstrate that such a system is able to successfully maintain different cell lines in culture for extended periods of time as well as significantly increasing throughput, accuracy, and reproducibility of HT and HC screens.     

RNAi living-cell microarrays for loss-of-function screens in Drosophila melanogaster cells

RNA interference (RNAi)-mediated loss-of-function screening in Drosophila melanogaster tissue culture cells is a powerful method for identifying the genes underlying cell biological functions and for annotating the fly genome. Here we describe the development of living-cell microarrays for screening large collections of RNAi-inducing double-stranded RNAs (dsRNAs) in Drosophila cells. The features of the microarrays consist of clusters of cells 200 mum in diameter, each with an RNAi-mediated depletion of a specific gene product. Because of the small size of the features, thousands of distinct dsRNAs can be screened on a single chip. The microarrays are suitable for quantitative and high-content cellular phenotyping and, in combination screens, for the identification of genetic suppressors, enhancers and synthetic lethal interactions. We used a prototype cell microarray with 384 different dsRNAs to identify previously unknown genes that affect cell proliferation and morphology, and, in a combination screen, that regulate dAkt/dPKB phosphorylation in the absence of dPTEN expression.     

Automatic Classification and Pattern Discovery in High-throughput Protein Crystallization Trials

Conceptually, protein crystallization can be divided into two phases search and optimization. Robotic protein crystallization screening can speed up the search phase, and has a potential to increase process quality. Automated image classification helps to increase throughput and consistently generate objective results. Although the classification accuracy can always be improved, our image analysis system can classify images from 1536-well plates with high classification accuracy (85%) and ROC score (0.87), as evaluated on 127 human-classified protein screens containing 5600 crystal images and 189472 non-crystal images. Data mining can integrate results from high-throughput screens with information about crystallizing conditions, intrinsic protein properties, and results from crystallization optimization. We apply association mining, a data mining approach that identifies frequently occurring patterns among variables and their values. This approach segregates proteins into groups based on how they react in a broad range of conditions, and clusters cocktails to reflect their potential to achieve crystallization. These results may lead to crystallization screen optimization, and reveal associations between protein properties and crystallization conditions. We also postulate that past experience may lead us to the identification of initial conditions favorable to crystallization for novel proteins.     

Functional assays for screening GPCR targets

G-protein-coupled receptors (GPCRs) are valuable molecular targets for drug discovery. An important aspect of the early drug discovery process is the design and implementation of high-throughput GPCR functional assays that allow the cost-effective screening of large compound libraries to identify novel drug candidates. Several functional assay kits based on fluorescence and/or chemiluminescence detection are commercially available for convenient screen development, each having advantages and disadvantages. In addition, new GPCR biosensors and high-content imaging technologies have recently been developed that hold promise for the development of functional GPCR screens in living cells.     

High-Content Analysis of Sequential Events during the Early Phase of Influenza A Virus Infection

Influenza A virus (IAV) represents a worldwide threat to public health by causing severe morbidity and mortality every year. Due to high mutation rate, new strains of IAV emerge frequently. These IAVs are often drug-resistant and require vaccine reformulation. A promising approach to circumvent this problem is to target host cell determinants crucial for IAV infection, but dispensable for the cell. Several RNAi-based screens have identified about one thousand cellular factors that promote IAV infection. However, systematic analyses to determine their specific functions are lacking. To address this issue, we developed quantitative, imaging-based assays to dissect seven consecutive steps in the early phases of IAV infection in tissue culture cells. The entry steps for which we developed the assays were: virus binding to the cell membrane, endocytosis, exposure to low pH in endocytic vacuoles, acid-activated fusion of viral envelope with the vacuolar membrane, nucleocapsid uncoating in the cytosol, nuclear import of viral ribonucleoproteins, and expression of the viral nucleoprotein. We adapted the assays to automated microscopy and optimized them for high-content screening. To quantify the image data, we performed both single and multi-parametric analyses, in combination with machine learning. By time-course experiments, we determined the optimal time points for each assay. Our quality control experiments showed that the assays were sufficiently robust for high-content analysis. The methods we describe in this study provide a powerful high-throughput platform to understand the host cell processes, which can eventually lead to the discovery of novel anti-pathogen strategies.     

Development of a high-content screening assay panel to accelerate mechanism of action studies for oncology research

Efficient elucidation of the biological mechanism of action of novel compounds remains a major bottleneck in the drug discovery process. To address this need in the area of oncology, we report the development of a multiparametric high-content screening assay panel at the level of single cells to dramatically accelerate understanding the mechanism of action of cell growth-inhibiting compounds on a large scale. Our approach is based on measuring 10 established end points associated with mitochondrial apoptosis, cell cycle disruption, DNA damage, and cellular morphological changes in the same experiment, across three multiparametric assays. The data from all of the measurements taken together are expected to help increase our current understanding of target protein functions, constrain the list of possible targets for compounds identified using phenotypic screens, and identify off-target effects. We have also developed novel data visualization and phenotypic classification approaches for detailed interpretation of individual compound effects and navigation of large collections of multiparametric cellular responses. We expect this general approach to be valuable for drug discovery across multiple therapeutic areas.     

Mammalian cell retention in a spinfilter perfusion bioreactor

A spinning cylindrical filter is often used to retain mammalian cells in a continuous perfusion bioreactor. This device, known as a spinfilter, has typically been with pore size smaller than the cell particles (single cells or aggregates) in order to achieve cell separation. For single cells in suspension, such an operation cannot be sustained over a long period of time because of clogging of the filter surface. Recently, screens with openings larger than the average cell size have been used to reduce the incidence of clogging. In this article, we have investigated how the screen size affects cell retention. We also showed why it is necessary to optimize the rotational speed of the spinfilter in order to achieve cell retention and reduce screen clogging. Effects of bulk mixing and perfusion rate on screen fouling cell retention, and cell washout were also investigated. (c) 1992 John Wiley & Sons, Inc.     

Identification of a small molecule that induces mitotic arrest using a simplified high-content screening assay and data analysis method

High-content screening has emerged as a new and powerful technique for identifying small-molecule modulators of mammalian cell biology. The authors describe the development and execution of a high-content screen to identify small molecules that induce mitotic arrest in mammalian cancer cells. Many widely used chemotherapeutics, such as Taxol and vinblastine, induce mitotic arrest, and the creation of new drugs that also induce mitotic arrest may have tremendous therapeutic value. In their screen, the authors employed a simple DNA stain (DAPI) and a sensitive nonparametric statistical test to identify compounds from an internal collection of approximately 13,000 high-quality lead-like small molecules. Subsequent analysis of 1 active compound indicated that it induces mitotic arrest, assessed using a high-content phosphohistone H3 detection assay, and caused cell proliferation defects in multiple cancer cell lines. The active compound, a quinazolinone originating from a natural product-like subset of the screened compounds, is active in cells at approximately 500 nM and appears to act by inhibiting the polymerization of tubulin.     

A high-content subtractive screen for selecting small molecules affecting internalization of GPCRs

G-protein-coupled receptors (GPCRs) are pivotal in cellular responses to the environment and are common drug targets. Identification of selective small molecules acting on single GPCRs is complicated by the shared machinery coupling signal transduction to physiology. Here, we demonstrate a high-content screen using a panel of GPCR assays to identify receptor selective molecules acting within the kinase/phosphatase inhibitor family. A collection of 88 kinase and phosphatase inhibitors was screened against seven agonist-induced GPCR internalization cell models as well as transferrin uptake in human embryonic kidney cells. Molecules acting on a single receptor were identified through excluding pan-specific compounds affecting housekeeping endocytosis or disrupting internalization of multiple receptors. We identified compounds acting on a sole GPCR from activities in a broad range of chemical structures that could not be easily sorted by conventional means. Selective analysis can therefore rapidly select compounds selectively affecting GPCR activity with specificity to one receptor class through high-content screening.     

Identification of RAS-mitogen-activated protein kinase signaling pathway modulators in an ERF1 redistribution screen

The RAS-mitogen-activated protein kinase (MAPK) signaling pathway has a central role in regulating the proliferation and survival of both normal and tumor cells. This pathway has been 1 focus area for the development of anticancer drugs, resulting in several compounds, primarily kinase inhibitors, in clinical testing. The authors have undertaken a cell-based, high-throughput screen using a novel ERF1 Redistribution assay to identify compounds that modulate the signaling pathway. The hit compounds were subsequently tested for activity in a functional cell proliferation assay designed to selectively detect compounds inhibiting the proliferation of MAPK pathway-dependent cancer cells. The authors report the identification of 2 cell membrane-permeable compounds that exhibit activity in the ERF1 Redistribution assay and selectively inhibit proliferation of MAPK pathway-dependent malignant melanoma cells at similar potencies (IC(50)=< 5 microM). These compounds have drug-like structures and are negative in RAF, MEK, and ERK in vitro kinase assays. Drugs belonging to these compound classes may prove useful for treating cancers caused by excessive MAPK pathway signaling. The results also show that cell-based, high-content Redistribution screens can detect compounds with different modes of action and reveal novel targets in a pathway known to be disease relevant.     

基于多特征融合的多分类运动想象脑电信号识别研究

针对目前多分类运动想象脑电识别存在特征提取单一、分类准确率低等问题,提出一种多特征融合的四分类运动想象脑电识别方法来提高识别率。对预处理后的脑电信号分别使用希尔伯特-黄变换、一对多共空间模式、近似熵、模糊熵、样本熵提取结合时频—空域—非线性动力学的初始特征向量,用主成分分析降维,最后使用粒子群优化支持向量机分类。该算法通过对国际标准数据集BCI2005 Data set IIIa中的k3b受试者数据经MATLAB仿真处理后获得93.30%的识别率,均高于单一特征和其它组合特征下的识别率。分别对四名实验者实验采集运动想象脑电数据,使用本研究提出的方法处理获得了72.96%的平均识别率。结果表明多特征融合的特征提取方法能更好的表征运动想象脑电信号,使用粒子群支持向量机可取得较高的识别准确率,为人脑的认知活动提供了一种新的识别方法。