Drug response profiles in patient-derived cancer cells across histological subtypes of ovarian cancer: real-time therapy tailoring for a patient with low-grade serous carcinoma.

Many efforts are underway to develop novel therapies against the aggressive high-grade serous ovarian cancers (HGSOCs), while our understanding of treatment options for low-grade (LGSOC) or mucinous (MUCOC) of ovarian malignancies is not developing as well. We describe here a functional precision oncology (fPO) strategy in epithelial ovarian cancers (EOC), which involves high-throughput drug testing of patient-derived ovarian cancer cells (PDCs) with a library of 526 oncology drugs, combined with genomic and transcriptomic profiling. HGSOC, LGSOC and MUCOC PDCs had statistically different overall drug response profiles, with LGSOCs responding better to targeted inhibitors than HGSOCs. We identified several subtype-specific drug responses, such as LGSOC PDCs showing high sensitivity to MDM2, ERBB2/EGFR inhibitors, MUCOC PDCs to MEK inhibitors, whereas HGSOCs showed strongest effects with CHK1 inhibitors and SMAC mimetics. We also explored several drug combinations and found that the dual inhibition of MEK and SHP2 was synergistic in MAPK-driven EOCs. We describe a clinical case study, where real-time fPO analysis of samples from a patient with metastatic, chemorefractory LGSOC with a CLU-NRG1 fusion guided clinical therapy selection. fPO-tailored therapy with afatinib, followed by trastuzumab and pertuzumab, successfully reduced tumour burden and blocked disease progression over a five-year period. In summary, fPO is a powerful approach for the identification of systematic drug response differences across EOC subtypes, as well as to highlight patient-specific drug regimens that could help to optimise therapies to individual patients in the future.

SARS-CoV-2-host proteome interactions for antiviral drug discovery.

Treatment options for COVID-19, caused by SARS-CoV-2, remain limited. Understanding viral pathogenesis at the molecular level is critical to develop effective therapy. Some recent studies have explored SARS-CoV-2-host interactomes and provided great resources for understanding viral replication. However, host proteins that functionally associate with SARS-CoV-2 are localized in the corresponding subnetwork within the comprehensive human interactome. Therefore, constructing a downstream network including all potential viral receptors, host cell proteases, and cofactors is necessary and should be used as an additional criterion for the validation of critical host machineries used for viral processing. This study applied both affinity purification mass spectrometry (AP-MS) and the complementary proximity-based labeling MS method (BioID-MS) on 29 viral ORFs and 18 host proteins with potential roles in viral replication to map the interactions relevant to viral processing. The analysis yields a list of 693 hub proteins sharing interactions with both viral baits and host baits and revealed their biological significance for SARS-CoV-2. Those hub proteins then served as a rational resource for drug repurposing via a virtual screening approach. The overall process resulted in the suggested repurposing of 59 compounds for 15 protein targets. Furthermore, antiviral effects of some candidate drugs were observed in vitro validation using image-based drug screen with infectious SARS-CoV-2. In addition, our results suggest that the antiviral activity of methotrexate could be associated with its inhibitory effect on specific protein-protein interactions.

iTReX: Interactive exploration of mono- and combination therapy dose response profiling data.

High throughput screening methods, measuring the sensitivity and resistance of tumor cells to drug treatments have been rapidly evolving. Not only do these screens allow correlating response profiles to tumor genomic features for developing novel predictors of treatment response, but they can also add evidence for therapy decision making in precision oncology. Recent analysis methods developed for either assessing single agents or combination drug efficacies enable quantification of dose-response curves with restricted symmetric fit settings. Here, we introduce iTReX, a user-friendly and interactive Shiny/R application, for both the analysis of mono- and combination therapy responses. The application features an extended version of the drug sensitivity score (DSS) based on the integral of an advanced five-parameter dose-response curve model and a differential DSS for combination therapy profiling. Additionally, iTReX includes modules that visualize drug target interaction networks and support the detection of matches between top therapy hits and the sample omics features to enable the identification of druggable targets and biomarkers. iTReX enables the analysis of various quantitative drug or therapy response readouts (e.g. luminescence, fluorescence microscopy) and multiple treatment strategies (drug treatments, radiation). Using iTReX we validate a cost-effective drug combination screening approach and reveal the application's ability to identify potential sample-specific biomarkers based on drug target interaction networks. The iTReX web application is accessible at https://itrex.kitz-heidelberg.de.

Loss of DIAPH1 causes SCBMS, combined immunodeficiency, and mitochondrial dysfunction.

BACKGROUND: Homozygous loss of DIAPH1 results in seizures, cortical blindness, and microcephaly syndrome (SCBMS). We studied 5 Finnish and 2 Omani patients with loss of DIAPH1 presenting with SCBMS, mitochondrial dysfunction, and immunodeficiency. OBJECTIVE: We sought to further characterize phenotypes and disease mechanisms associated with loss of DIAPH1. METHODS: Exome sequencing, genotyping and haplotype analysis, B- and T-cell phenotyping, in vitro lymphocyte stimulation assays, analyses of mitochondrial function, immunofluorescence staining for cytoskeletal proteins and mitochondria, and CRISPR-Cas9 DIAPH1 knockout in heathy donor PBMCs were used. RESULTS: Genetic analyses found all Finnish patients homozygous for a rare DIAPH1 splice-variant (NM_005219:c.684+1G>A) enriched in the Finnish population, and Omani patients homozygous for a previously described pathogenic DIAPH1 frameshift-variant (NM_005219:c.2769delT;p.F923fs). In addition to microcephaly, epilepsy, and cortical blindness characteristic to SCBMS, the patients presented with infection susceptibility due to defective lymphocyte maturation and 3 patients developed B-cell lymphoma. Patients' immunophenotype was characterized by poor lymphocyte activation and proliferation, defective B-cell maturation, and lack of naive T cells. CRISPR-Cas9 knockout of DIAPH1 in PBMCs from healthy donors replicated the T-cell activation defect. Patient-derived peripheral blood T cells exhibited impaired adhesion and inefficient microtubule-organizing center repositioning to the immunologic synapse. The clinical symptoms and laboratory tests also suggested mitochondrial dysfunction. Experiments with immortalized, patient-derived fibroblasts indicated that DIAPH1 affects the amount of complex IV of the mitochondrial respiratory chain. CONCLUSIONS: Our data demonstrate that individuals with SCBMS can have combined immune deficiency and implicate defective cytoskeletal organization and mitochondrial dysfunction in SCBMS pathogenesis.

Association of tamoxifen resistance and lipid reprogramming in breast cancer.

BACKGROUND: Tamoxifen treatment of estrogen receptor (ER)-positive breast cancer reduces mortality by 31%. However, over half of advanced ER-positive breast cancers are intrinsically resistant to tamoxifen and about 40% will acquire the resistance during the treatment. METHODS: In order to explore mechanisms underlying endocrine therapy resistance in breast cancer and to identify new therapeutic opportunities, we created tamoxifen-resistant breast cancer cell lines that represent the luminal A or the luminal B. Gene expression patterns revealed by RNA-sequencing in seven tamoxifen-resistant variants were compared with their isogenic parental cells. We further examined those transcriptomic alterations in a publicly available patient cohort. RESULTS: We show that tamoxifen resistance cannot simply be explained by altered expression of individual genes, common mechanism across all resistant variants, or the appearance of new fusion genes. Instead, the resistant cell lines shared altered gene expression patterns associated with cell cycle, protein modification and metabolism, especially with the cholesterol pathway. In the tamoxifen-resistant T-47D cell variants we observed a striking increase of neutral lipids in lipid droplets as well as an accumulation of free cholesterol in the lysosomes. Tamoxifen-resistant cells were also less prone to lysosomal membrane permeabilization (LMP) and not vulnerable to compounds targeting the lipid metabolism. However, the cells were sensitive to disulfiram, LCS-1, and dasatinib. CONCLUSION: Altogether, our findings highlight a major role of LMP prevention in tamoxifen resistance, and suggest novel drug vulnerabilities associated with this phenotype.

NDRG1 functions in LDL receptor trafficking by regulating endosomal recycling and degradation.

N-myc downstream-regulated gene 1 (NDRG1) mutations cause Charcot-Marie-Tooth disease type 4D (CMT4D). However, the cellular function of NDRG1 and how it causes CMT4D are poorly understood. We report that NDRG1 silencing in epithelial cells results in decreased uptake of low-density lipoprotein (LDL) due to reduced LDL receptor (LDLR) abundance at the plasma membrane. This is accompanied by the accumulation of LDLR in enlarged EEA1-positive endosomes that contain numerous intraluminal vesicles and sequester ceramide. Concomitantly, LDLR ubiquitylation is increased but its degradation is reduced and ESCRT (endosomal sorting complex required for transport) proteins are downregulated. Co-depletion of IDOL (inducible degrader of the LDLR), which ubiquitylates the LDLR and promotes its degradation, rescues plasma membrane LDLR levels and LDL uptake. In murine oligodendrocytes, Ndrg1 silencing not only results in reduced LDL uptake but also in downregulation of the oligodendrocyte differentiation factor Olig2. Both phenotypes are rescued by co-silencing of Idol, suggesting that ligand uptake through LDLR family members controls oligodendrocyte differentiation. These findings identify NDRG1 as a novel regulator of multivesicular body formation and endosomal LDLR trafficking. The deficiency of functional NDRG1 in CMT4D might impair lipid processing and differentiation of myelinating cells.

Drug repurposing platform for deciphering the druggable SARS-CoV-2 interactome.

The coronavirus disease 2019 (COVID-19) pandemic has heavily challenged the global healthcare system. Despite the vaccination programs, the new virus variants are circulating. Further research is required for understanding of the biology of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and for discovery of therapeutic agents against the virus. Here, we took advantage of drug repurposing to identify if existing drugs could inhibit SARS-CoV-2 infection. We established an open high throughput platform for in vitro screening of drugs against SARS-CoV-2 infection. We screened ∼1000 drugs for their ability to inhibit SARS-CoV-2-induced cell death in the African green monkey kidney cell line (Vero-E6), analyzed how the hit compounds affect the viral N (nucleocapsid) protein expression in human cell lines using high-content microscopic imaging and analysis, determined the hit drug targets in silico, and assessed their ability to cause phospholipidosis, which can interfere with the viral replication. Duvelisib was found by in silico interaction assay as a potential drug targeting virus-host protein interactions. The predicted interaction between PARP1 and S protein, affected by Duvelisib, was further validated by immunoprecipitation. Our results represent a rapidly applicable platform for drug repurposing and evaluation of the new emerging viruses' responses to the drugs. Further in silico studies help us to discover the druggable host pathways involved in the infectious cycle of SARS-CoV-2.

Protocol for 3D drug sensitivity and resistance testing of patient-derived cancer cells in 384-well plates.

Establishment of drug testing of patient-derived cancer cells (PDCs) in physiologically relevant 3-dimensional (3D) culture is central for drug discovery and cancer research, as well as for functional precision medicine. Here, we describe the detailed protocol allowing the 3D drug testing of PDCs - or any type of cells of interest - in Matrigel in 384-well plate format using automation. We also provide an alternative protocol, which does not require supporting matrices. The cancer tissue is obtained directly from clinics (after surgery or biopsy) and processed into single cell suspension. Systematic drug sensitivity and resistance testing (DSRT) is carried out on the PDCs directly after cancer cell isolation from tissue or on cells expanded for a few passages. In the 3D-DSRT assay, the PDCs are plated in 384-well plates in Matrigel, grown as spheroids, and treated with compounds of interest for 72 h. The cell viability is directly measured using a luminescence-based assay. Alternatively, prior to the cell viability measurement, drug-treated cells can be directly subjected to automated high-content bright field imaging or stained for fluorescence (live) cell microscopy for further image analysis. This is followed by the quality control and data analysis. The 3D-DSRT can be performed within a 1-3-week timeframe of the clinical sampling of cancer tissue, depending on the amount of the obtained tissue, growth rate of cancer cells, and the number of drugs being tested. The 3D-DSRT method can be flexibly modified, e.g., to be carried out with or without supporting matrices with U-bottom 384-well plates when appropriate for the PDCs or other cell models used.

Comparison of two supporting matrices for patient-derived cancer cells in 3D drug sensitivity and resistance testing assay (3D-DSRT).

Central to the success of functional precision medicine of solid tumors is to perform drug testing of patient-derived cancer cells (PDCs) in tumor-mimicking ex vivo conditions. While high throughput (HT) drug screening methods have been well-established for cells cultured in two-dimensional (2D) format, this approach may have limited value in predicting clinical responses. Here, we describe the results of the optimization of drug sensitivity and resistance testing (DSRT) in three-dimensional (3D) growth supporting matrices in a HT mode (3D-DSRT) using the hepatocyte cell line (HepG2) as an example. Supporting matrices included widely used animal-derived Matrigel and cellulose-based hydrogel, GrowDex, which has earlier been shown to support 3D growth of cell lines and stem cells. Further, the sensitivity of ovarian cancer PDCs, from two patients included in the functional precision medicine study, was tested for 52 drugs in 5 different concentrations using 3D-DSRT. Shortly, in the optimized protocol, the PDCs are embedded with matrices and seeded to 384-well plates to allow the formation of the spheroids prior to the addition of drugs in nanoliter volumes with acoustic dispenser. The sensitivity of spheroids to drug treatments is measured with cell viability readout (here, 72 h after addition of drugs). The quality control and data analysis are performed with openly available Breeze software. We show the usability of both matrices in established 3D-DSRT, and report 2D vs 3D growth condition dependent differences in sensitivities of ovarian cancer PDCs to MEK-inhibitors and cytotoxic drugs. This study provides a proof-of-concept for robust and fast screening of drug sensitivities of PDCs in 3D-DSRT, which is important not only for drug discovery but also for personalized ex vivo drug testing in functional precision medicine studies. These findings suggest that comparing results of 2D- and 3D-DSRT is essential for understanding drug mechanisms and for selecting the most effective treatment for the patient.

Ex Vivo Drug Testing in Patient-derived Papillary Renal Cancer Cells Reveals EGFR and the BCL2 Family as Therapeutic Targets.

BACKGROUND: Immune checkpoint inhibitors and antiangiogenic agents are used for first-line treatment of advanced papillary renal cell carcinoma (pRCC) but pRCC response rates to these therapies are low. OBJECTIVE: To generate and characterise a functional ex vivo model to identify novel treatment options in advanced pRCC. DESIGN, SETTING, AND PARTICIPANTS: We established patient-derived cell cultures (PDCs) from seven pRCC samples from patients and characterised them via genomic analysis and drug profiling. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Comprehensive molecular characterisation in terms of copy number analysis and whole-exome sequencing confirmed the concordance of pRCC PDCs with the original tumours. We evaluated their sensitivity to novel drugs by generating drug scores for each PDC. RESULTS AND LIMITATIONS: PDCs confirmed pRCC-specific copy number variations such as gains in chromosomes 7, 16, and 17. Whole-exome sequencing revealed that PDCs retained mutations in pRCC-specific driver genes. We performed drug screening with 526 novel and oncological compounds. Whereas exposure to conventional drugs showed low efficacy, the results highlighted EGFR and BCL2 family inhibition as the most effective targets in our pRCC PDCs. CONCLUSIONS: High-throughput drug testing on newly established pRCC PDCs revealed that inhibition of EGFR and BCL2 family members could be a therapeutic strategy in pRCC. PATIENT SUMMARY: We used a new approach to generate patient-derived cells from a specific type of kidney cancer. We showed that these cells have the same genetic background as the original tumour and can be used as models to study novel treatment options for this type of kidney cancer.

Image-based and machine learning-guided multiplexed serology test for SARS-CoV-2.

We present a miniaturized immunofluorescence assay (mini-IFA) for measuring antibody response in patient blood samples. The method utilizes machine learning-guided image analysis and enables simultaneous measurement of immunoglobulin M (IgM), IgA, and IgG responses against different viral antigens in an automated and high-throughput manner. The assay relies on antigens expressed through transfection, enabling use at a low biosafety level and fast adaptation to emerging pathogens. Using severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) as the model pathogen, we demonstrate that this method allows differentiation between vaccine-induced and infection-induced antibody responses. Additionally, we established a dedicated web page for quantitative visualization of sample-specific results and their distribution, comparing them with controls and other samples. Our results provide a proof of concept for the approach, demonstrating fast and accurate measurement of antibody responses in a research setup with prospects for clinical diagnostics.

Drug sensitivity profiling of 3D tumor tissue cultures in the pediatric precision oncology program INFORM.

The international precision oncology program INFORM enrolls relapsed/refractory pediatric cancer patients for comprehensive molecular analysis. We report a two-year pilot study implementing ex vivo drug sensitivity profiling (DSP) using a library of 75-78 clinically relevant drugs. We included 132 viable tumor samples from 35 pediatric oncology centers in seven countries. DSP was conducted on multicellular fresh tumor tissue spheroid cultures in 384-well plates with an overall mean processing time of three weeks. In 89 cases (67%), sufficient viable tissue was received; 69 (78%) passed internal quality controls. The DSP results matched the identified molecular targets, including BRAF, ALK, MET, and TP53 status. Drug vulnerabilities were identified in 80% of cases lacking actionable (very) high-evidence molecular events, adding value to the molecular data. Striking parallels between clinical courses and the DSP results were observed in selected patients. Overall, DSP in clinical real-time is feasible in international multicenter precision oncology programs.

High-Content Imaging to Phenotype Human Primary and iPSC-Derived Cells.

Increasingly powerful microscopy, liquid handling, and computational techniques have enabled cell imaging in high throughput. Microscopy images are quantified using high-content analysis platforms linking object features to cell behavior. This can be attempted on physiologically relevant cell models, including stem cells and primary cells, in complex environments, and conceivably in the presence of perturbations. Recently, substantial focus has been devoted to cell profiling for cell therapy, assays for drug discovery or biomarker identification for clinical decision-making protocols, bringing this wealth of information into translational applications. In this chapter, we focus on two protocols enabling to (1) benchmark human cells, in particular human endothelial cells as a case study and (2) extract cells from blood for follow-up experiments including image-based drug testing. We also present concepts of high-content imaging and discuss the benefits and challenges, with the aim of enabling readers to tailor existing pipelines and bring such approaches closer to translational research and the clinic.

SpheroidPicker for automated 3D cell culture manipulation using deep learning.

Recent statistics report that more than 3.7 million new cases of cancer occur in Europe yearly, and the disease accounts for approximately 20% of all deaths. High-throughput screening of cancer cell cultures has dominated the search for novel, effective anticancer therapies in the past decades. Recently, functional assays with patient-derived ex vivo 3D cell culture have gained importance for drug discovery and precision medicine. We recently evaluated the major advancements and needs for the 3D cell culture screening, and concluded that strictly standardized and robust sample preparation is the most desired development. Here we propose an artificial intelligence-guided low-cost 3D cell culture delivery system. It consists of a light microscope, a micromanipulator, a syringe pump, and a controller computer. The system performs morphology-based feature analysis on spheroids and can select uniform sized or shaped spheroids to transfer them between various sample holders. It can select the samples from standard sample holders, including Petri dishes and microwell plates, and then transfer them to a variety of holders up to 384 well plates. The device performs reliable semi- and fully automated spheroid transfer. This results in highly controlled experimental conditions and eliminates non-trivial side effects of sample variability that is a key aspect towards next-generation precision medicine.

Regression plane concept for analysing continuous cellular processes with machine learning.

Biological processes are inherently continuous, and the chance of phenotypic discovery is significantly restricted by discretising them. Using multi-parametric active regression we introduce the Regression Plane (RP), a user-friendly discovery tool enabling class-free phenotypic supervised machine learning, to describe and explore biological data in a continuous manner. First, we compare traditional classification with regression in a simulated experimental setup. Second, we use our framework to identify genes involved in regulating triglyceride levels in human cells. Subsequently, we analyse a time-lapse dataset on mitosis to demonstrate that the proposed methodology is capable of modelling complex processes at infinite resolution. Finally, we show that hemocyte differentiation in Drosophila melanogaster has continuous characteristics.

PeptiCHIP: A Microfluidic Platform for Tumor Antigen Landscape Identification.

Identification of HLA class I ligands from the tumor surface (ligandome or immunopeptidome) is essential for designing T-cell mediated cancer therapeutic approaches. However, the sensitivity of the process for isolating MHC-I restricted tumor-specific peptides has been the major limiting factor for reliable tumor antigen characterization, making clear the need for technical improvement. Here, we describe our work from the fabrication and development of a microfluidic-based chip (PeptiCHIP) and its use to identify and characterize tumor-specific ligands on clinically relevant human samples. Specifically, we assessed the potential of immobilizing a pan-HLA antibody on solid surfaces via well-characterized streptavidin-biotin chemistry, overcoming the limitations of the cross-linking chemistry used to prepare the affinity matrix with the desired antibodies in the immunopeptidomics workflow. Furthermore, to address the restrictions related to the handling and the limited availability of tumor samples, we further developed the concept toward the implementation of a microfluidic through-flow system. Thus, the biotinylated pan-HLA antibody was immobilized on streptavidin-functionalized surfaces, and immune-affinity purification (IP) was carried out on customized microfluidic pillar arrays made of thiol-ene polymer. Compared to the standard methods reported in the field, our methodology reduces the amount of antibody and the time required for peptide isolation. In this work, we carefully examined the specificity and robustness of our customized technology for immunopeptidomics workflows. We tested this platform by immunopurifying HLA-I complexes from 1 × 106 cells both in a widely studied B-cell line and in patients-derived ex vivo cell cultures, instead of 5 × 108 cells as required in the current technology. After the final elution in mild acid, HLA-I-presented peptides were identified by tandem mass spectrometry and further investigated by in vitro methods. These results highlight the potential to exploit microfluidics-based strategies in immunopeptidomics platforms and in personalized immunopeptidome analysis from cells isolated from individual tumor biopsies to design tailored cancer therapeutic vaccines. Moreover, the possibility to integrate multiple identical units on a single chip further improves the throughput and multiplexing of these assays with a view to clinical needs.

Rab8 regulates ABCA1 cell surface expression and facilitates cholesterol efflux in primary human macrophages.

OBJECTIVE: ATP-binding cassette transporter A1 (ABCA1) is thought to lipidate apolipoprotein A-I (apoA-I) at the plasma membrane, with endosomal cholesterol contributing as substrate. The mechanisms of ABCA1 surface delivery are not well understood. We have shown that Rab8 regulates endosomal cholesterol removal to apoA-I in human fibroblasts. Here, we investigated whether Rab8 plays a role in ABCA1 plasma membrane expression and cholesterol removal in primary human macrophages. METHODS AND RESULTS: We found that Rab8 was abundantly expressed in human atherosclerotic lesional macrophages and upregulated on lipid loading of macrophages in vitro. Adenoviral overexpression of Rab8 increased ABCA1 protein levels and reduced cholesterol deposition in macrophage foam cells incubated with apoA-I. Depletion of Rab8 decreased the fraction of ABCA1 at the plasma membrane and inhibited the efflux of lipoprotein-derived endosomal cholesterol to apoA-I. In Rab8-depleted cells, ABCA1-GFP localized in beta1 integrin and transferrin receptor containing recycling organelles. CONCLUSIONS: Rab8 reduces foam cell formation by facilitating ABCA1 surface expression and stimulating endosomal cholesterol efflux to apoA-I in primary human macrophages.

Echovirus 1 endocytosis into caveosomes requires lipid rafts, dynamin II, and signaling events.

Binding of echovirus 1 (EV1, a nonenveloped RNA virus) to the alpha2beta1 integrin on the cell surface is followed by endocytic internalization of the virus together with the receptor. Here, video-enhanced live microscopy revealed the rapid uptake of fluorescently labeled EV1 into mobile, intracellular structures, positive for green fluorescent protein-tagged caveolin-1. Partial colocalization of EV1 with SV40 (SV40) and cholera toxin, known to traffic via caveosomes, demonstrated that the vesicles were caveosomes. The initiation of EV1 infection was dependent on dynamin II, cholesterol, and protein phosphorylation events. Brefeldin A, a drug that prevents SV40 transport, blocked the EV1 infection cycle, whereas drugs that disrupt the cellular cytoskeleton had no effect. In situ hybridization revealed the localization of viral RNA with endocytosed viral capsid proteins in caveosomes before initiation of viral replication. Thus, both the internalization of EV1 to caveosomes and subsequent events differ clearly from caveolar endocytosis of SV40 because EV1 uptake is fast and independent of actin and EV1 is not sorted further to sER from caveosomes. These results shed further light on the cell entry of nonenveloped viral pathogens and illustrate the use of viruses as probes to dissect caveolin-associated endocytic pathways.

Advanced Cell Classifier: User-Friendly Machine-Learning-Based Software for Discovering Phenotypes in High-Content Imaging Data.

High-content, imaging-based screens now routinely generate data on a scale that precludes manual verification and interrogation. Software applying machine learning has become an essential tool to automate analysis, but these methods require annotated examples to learn from. Efficiently exploring large datasets to find relevant examples remains a challenging bottleneck. Here, we present Advanced Cell Classifier (ACC), a graphical software package for phenotypic analysis that addresses these difficulties. ACC applies machine-learning and image-analysis methods to high-content data generated by large-scale, cell-based experiments. It features methods to mine microscopic image data, discover new phenotypes, and improve recognition performance. We demonstrate that these features substantially expedite the training process, successfully uncover rare phenotypes, and improve the accuracy of the analysis. ACC is extensively documented, designed to be user-friendly for researchers without machine-learning expertise, and distributed as a free open-source tool at www.cellclassifier.org.

Accurate Morphology Preserving Segmentation of Overlapping Cells based on Active Contours.

The identification of fluorescently stained cell nuclei is the basis of cell detection, segmentation, and feature extraction in high content microscopy experiments. The nuclear morphology of single cells is also one of the essential indicators of phenotypic variation. However, the cells used in experiments can lose their contact inhibition, and can therefore pile up on top of each other, making the detection of single cells extremely challenging using current segmentation methods. The model we present here can detect cell nuclei and their morphology even in high-confluency cell cultures with many overlapping cell nuclei. We combine the "gas of near circles" active contour model, which favors circular shapes but allows slight variations around them, with a new data model. This captures a common property of many microscopic imaging techniques: the intensities from superposed nuclei are additive, so that two overlapping nuclei, for example, have a total intensity that is approximately double the intensity of a single nucleus. We demonstrate the power of our method on microscopic images of cells, comparing the results with those obtained from a widely used approach, and with manual image segmentations by experts.