An integrated one-step system to extract, analyze and annotate all relevant information from image-based cell screening of chemical libraries.

Here we report the development and validation of a complete solution to manage and analyze the data produced by image-based phenotypic screening campaigns of small-molecule libraries. In one step initial crude images are analyzed for multiple cytological features, statistical analysis is performed and molecules that produce the desired phenotypic profile are identified. A naïve Bayes classifier, integrating chemical and phenotypic spaces, is built and utilized during the process to assess those images initially classified as "fuzzy"-an automated iterative feedback tuning. Simultaneously, all this information is directly annotated in a relational database containing the chemical data. This novel fully automated method was validated by conducting a re-analysis of results from a high-content screening campaign involving 33 992 molecules used to identify inhibitors of the PI3K/Akt signaling pathway. Ninety-two percent of confirmed hits identified by the conventional multistep analysis method were identified using this integrated one-step system as well as 40 new hits, 14.9% of the total, originally false negatives. Ninety-six percent of true negatives were properly recognized too. A web-based access to the database, with customizable data retrieval and visualization tools, facilitates the posterior analysis of annotated cytological features which allows identification of additional phenotypic profiles; thus, further analysis of original crude images is not required.

[1]  Robert A Blake,et al.  Target validation in drug discovery. , 2007, Methods in molecular biology.

[2]  P. Willett,et al.  Comparison of topological descriptors for similarity-based virtual screening using multiple bioactive reference structures. , 2004, Organic & biomolecular chemistry.

[3]  M. Sipe,et al.  Detection and classification of threat agents via high‐content assays of mammalian cells , 2004, Journal of applied toxicology : JAT.

[4]  Pamela A. Silver,et al.  A chemical genetic screen identifies inhibitors of regulated nuclear export of a Forkhead transcription factor in PTEN-deficient tumor cells. , 2003, Cancer cell.

[5]  D. L. Taylor,et al.  Advances in high content screening for drug discovery. , 2003, Assay and drug development technologies.

[6]  Andreas Bender,et al.  Molecular Similarity Searching Using Atom Environments, Information-Based Feature Selection, and a Naïve Bayesian Classifier , 2004, J. Chem. Inf. Model..

[7]  Pier Paolo Pandolfi,et al.  The Multiple Roles of PTEN in Tumor Suppression , 2000, Cell.

[8]  Xiaoyang Xia,et al.  Classification of kinase inhibitors using a Bayesian model. , 2004, Journal of medicinal chemistry.

[9]  Polina Golland,et al.  Scoring diverse cellular morphologies in image-based screens with iterative feedback and machine learning , 2009, Proceedings of the National Academy of Sciences.

[10]  M. Eisenstein,et al.  Automated imaging: data as far as the eye can see , 2005, Nature Methods.

[11]  Wolfgang Link,et al.  Chemical Genetic Analysis of FOXO Nuclear–Cytoplasmic Shuttling by Using Image‐Based Cell Screening , 2008, Chembiochem : a European journal of chemical biology.

[12]  Obdulia Rabal,et al.  3-Kinases Selective Inhibitors of Phosphoinositide andNuclear Translocation Identifies Potent Chemical Interrogation of FOXO 3 a Mechanisms of Signal Transduction : , 2009 .

[13]  Timothy J Mitchison,et al.  Small‐Molecule Screening and Profiling by Using Automated Microscopy , 2005, Chembiochem : a European journal of chemical biology.

[14]  K. Yeow,et al.  Cellular imaging in drug discovery , 2006, Nature Reviews Drug Discovery.

[15]  B. Howell,et al.  High-content screening: emerging hardware and software technologies. , 2006, Methods in enzymology.

[16]  Anne E Carpenter.  Image-based chemical screening. , 2007, Nature chemical biology.

[17]  Timothy J Mitchison,et al.  Small molecule screening by imaging. , 2006, Current opinion in chemical biology.

[18]  John Bell,et al.  A review of methods for the assessment of prediction errors in conservation presence/absence models , 1997, Environmental Conservation.

[19]  J. Waring,et al.  Application of a High-Content Multiparameter Cytotoxicity Assay to Prioritize Compounds Based on Toxicity Potential in Humans , 2008, Journal of biomolecular screening.

[20]  D. Rogers,et al.  Using Extended-Connectivity Fingerprints with Laplacian-Modified Bayesian Analysis in High-Throughput Screening Follow-Up , 2005, Journal of biomolecular screening.

[21]  Albert Gough,et al.  High-Content Screening: A New Approach to Easing Key Bottlenecks in the Drug Discovery Process , 1997 .

[22]  Anne E Carpenter.  Software opens the door to quantitative imaging , 2007, Nature Methods.

[23]  Meir Glick,et al.  Enrichment of Extremely Noisy High-Throughput Screening Data Using a Naïve Bayes Classifier , 2004, Journal of biomolecular screening.

[24]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[25]  Stuart L Schreiber,et al.  Identification of a small-molecule inhibitor of class Ia PI3Ks with cell-based screening. , 2007, Chemistry & biology.

[26]  R. Ricciardi,et al.  Identification of non-nucleoside DNA synthesis inhibitors of vaccinia virus by high-throughput screening. , 2008, Journal of medicinal chemistry.