Spatial omics techniques and data analysis for cancer immunotherapy applications

In-depth profiling of cancer cells/tissues is expanding our understanding of the genomic, epigenomic, transcriptomic, and proteomic landscape of cancer. However, the complexity of the cancer microenvironment, particularly its immune regulation, has made it difficult to exploit the potential of cancer immunotherapy. High-throughput spatial omics technologies and analysis pipelines have emerged as powerful tools for tackling this challenge. As a result, a potential revolution in cancer diagnosis, prognosis, and treatment is on the horizon. In this review, we discuss the technological advances in spatial profiling of cancer around and beyond the central dogma to harness the full benefits of immunotherapy. We also discuss the promise and challenges of spatial data analysis and interpretation and provide an outlook for the future.

Global Epidemiology and Genetics of Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is one of the leading cancers worldwide. Classically, HCC develops in genetically susceptible individuals who are exposed to risk factors, especially in the presence of liver cirrhosis. Significant temporal and geographic variations exist for HCC and its etiologies. Over time, the burden of HCC has shifted from the low-moderate to the high sociodemographic index regions, reflecting the transition from viral to nonviral causes. Geographically, the hepatitis viruses predominate as the causes of HCC in Asia and Africa. Although there are genetic conditions that confer increased risk for HCC, these diagnoses are rarely recognized outside North America and Europe. In this review, we will evaluate the epidemiologic trends and risk factors of HCC, and discuss the genetics of HCC, including monogenic diseases, single-nucleotide polymorphisms, gut microbiome, and somatic mutations.

Choice of PD-L1 immunohistochemistry assay influences clinical eligibility for gastric cancer immunotherapy

4026

Background: Immune checkpoint inhibitors (ICI) are now standard-of-care treatment for patients with metastatic gastric cancer (GC). To guide patient selection for ICI therapy, programmed death ligand-1 (PD-L1) biomarker expression is routinely assessed via immunohistochemistry (IHC). Regulatory approval for ICIs is granted based on PD-L1 expression status, scored using metrics such as the combined positive score (CPS). However, with an increasing number of approved ICIs, each paired with a different PD-L1 antibody IHC assay used in their respective landmark trials, there is an unmet clinical and logistical need for harmonization. We thus investigated the interchangeability between the Dako 22C3, Dako 28-8 and Ventana SP-142 assays in GC PD-L1 IHC. Methods: In this cross-sectional study, samples were obtained via biopsy or resection of gastric cancer at the National University Hospital, Singapore. We scored 362 GC samples for PD-L1 CPS, tumor proportion score (TPS) and immune cells (IC) using a multiplex immunohistochemistry/immunofluorescence technique. 344 samples were developed into a tissue microarray (TMA), while 18 samples were used as whole slides for orthogonal validation. The samples selected for whole slide analysis were obtained from GC patients treated with ICI therapy. Results: The percentage of PD-L1 positive samples at clinically relevant CPS ≥1, ≥5 and ≥10 cut-offs (Table) for the 28-8 assay were approximately two-fold higher than that of the 22C3 (CPS≥1: 70.3% vs 49.4%, p<0.001; CPS≥5: 29.1% vs 13.4%, p<0.001; CPS≥10: 13.7% vs 7.0%, p=0.004). The mean CPS score on 28-8 assay was nearly double that of the 22C3 (6.39 ±14.5 vs 3.46±8.98, p<0.001). At the clinically important CPS≥5 cut-off, there was only moderate concordance between the 22C3 and 28-8 assays. Conclusions: Our findings suggest that scoring PD-L1 CPS with the 28-8 assay may result in higher proportion of PD-L1 positivity and higher PD-L1 scores compared to assessment with the 22C3 and other assays. Clinically, this could lead to a larger number of patients eligible and approved for ICI therapy. If assays are viewed and used interchangeability, a substantial number of patients may be inaccurately denied or granted treatment with ICIs based on the assay chosen. As such, until stronger evidence of inter-assay concordance is found, we urge caution in treating the assays as equivalent.[Table: see text]

Choice of PD-L1 immunohistochemistry assay influences clinical eligibility for gastric cancer immunotherapy

BACKGROUND

Immune checkpoint inhibitors (ICI) are now standard-of-care treatment for patients with metastatic gastric cancer (GC). To guide patient selection for ICI therapy, programmed death ligand-1 (PD-L1) biomarker expression is routinely assessed via immunohistochemistry (IHC). However, with an increasing number of approved ICIs, each paired with a different PD-L1 antibody IHC assay used in their respective landmark trials, there is an unmet clinical and logistical need for harmonization. We investigated the interchangeability between the Dako 22C3, Dako 28-8 and Ventana SP-142 assays in GC PD-L1 IHC.

METHODS

In this cross-sectional study, we scored 362 GC samples for PD-L1 combined positive score (CPS), tumor proportion score (TPS) and immune cells (IC) using a multiplex immunohistochemistry/immunofluorescence technique. Samples were obtained via biopsy or resection of gastric cancer.

RESULTS

The percentage of PD-L1-positive samples at clinically relevant CPS ≥ 1, ≥ 5 and ≥ 10 cut-offs for the 28-8 assay were approximately two-fold higher than that of the 22C3 (CPS ≥ 1: 70.3 vs 49.4%, p < 0.001; CPS ≥ 5: 29.1 vs 13.4%, p < 0.001; CPS ≥ 10: 13.7 vs 7.0%, p = 0.004). The mean CPS score on 28-8 assay was nearly double that of the 22C3 (6.39 ± 14.5 vs 3.46 ± 8.98, p < 0.001). At the clinically important CPS ≥ 5 cut-off, there was only moderate concordance between the 22C3 and 28-8 assays.

CONCLUSION

Our findings suggest that scoring PD-L1 CPS with the 28-8 assay may result in higher PD-L1 scores and higher proportion of PD-L1 positivity compared to 22C3 and other assays. Until stronger evidence of inter-assay concordance is found, we urge caution in treating the assays as equivalent.

Group VIII Metal Carbonyl Cluster-Boronic Acid Conjugates: Cytotoxicity and Mode of Action Studies

A set of metal carbonyl cluster-boronic acid conjugates of the group VIII metals (Fe, Ru, and Os) were synthesized and their antiproliferative effects measured against two breast cancer cell lines (MCF-7 and MDA-MB-231) and a noncancerous breast epithelial (MCF-10A) cell line. The cytotoxicity followed the order Ru > Os > Fe for the MDA-MB-231 cells, although the latter two exhibited similar cytotoxicity against MCF-7 and MCF-10A cells. The osmium species {Os₃(CO)₁₀(μ-H)[μ-SC₆H₄-p-B(OH)₂]} (2) could be chemically oxidized to its hydroxy analogue [Os₃(CO)₁₀(μ-H)(μ-SC₆H₄ -p-OH)] (2-OH), which showed comparable cytotoxicity. Mode of action studies pointed to an apoptotic pathway for cell death.

627 ImmunoAtlas: an online public portal for sharing, visualizing, and referencing multiplex immunohistochemistry/immunofluorescence (mIHC/IF) images and results for immuno-oncology

Background

Recent advances in multiplex immunohistochemistry/immunofluorescence (mIHC/IF) technologies have enabled simultaneous measurements of large numbers of markers on the same tissue sections, and more comprehensive views of the cellular compositions and immune responses of the tumor microenvironment. However, the reproducibility and interpretation of the complex staining patterns and analysis results obtained from these markers remain major barriers to more general adoptions of these technologies. Here, we report the availability of an online public portal, “ImmunoAtlas”, which would enable researchers/clinicians to present or share their published mIHC/IF images or results; international workgroups to create and share standard marker panels or assay guidelines; end users to validate antibodies or protocols; or computational scientists to benchmark different analysis methods on standard reference images (figure 1).

Methods

ImmunoAtlas is based on a HistoPathological image management and Analysis (HPA) software platform developed by us, and hosted in the data center of Bioinformatics Institute. The platform uses the cellXpress software,1 which is written in C++ and supports parallel processing, to efficiently process and manage large numbers of huge mIHC/IF or brightfield images. The web interface of ImmunoAtlas is also completely developed by us in PHP and JavaScript to address the specific needs and requirements in managing these images.

Results

ImmunoAtlas is a user-friendly online portal for sharing, visualizing, and referencing original tissue images and analysis results (https://ImmunoAtlas.org). We have completed the first phase of development of the portal. Users can now perform image uploading, annotation, publishing, sharing, and viewing with standard web browsers on desktop computers or mobile devices/phones (figure 2). The portal supports image files from common microscopes and slide scanners, and can process mIHC/IF or brightfield images from selected areas, tissues microarrays, or whole slides. It can handle up to 1000 multiplexed markers, and whole-slide images that are >20GB/image. Several internal and external immuno-oncology studies have deposited and shared their images via ImmunoAtlas. They include a study of multiplexed PD-L1 markers in breast cancers2; the development of a panel of 56 highly-multiplexed markers for cutaneous T cell lymphoma3; and a study of CD38 scoring in hepatocellular carcinoma.4

Conclusions

ImmunoAtlas promotes open science and collaborations that can accelerate the adoptions of mIHC/IF technologies in immuno-oncology. The next phase of development will focus on adding image searching and comparison functions to the portal. The community is welcome to use and share their images and analysis results via the portal.

References

Laksameethanasan D, Tan RZ, Toh GW-L, et al. cellXpress: a fast and user-friendly software platform for profiling cellular phenotypes. BMC Bioinformatics 2013;14:S4. doi:10.1186/1471-2105-14-S16-S4.

Yeong J, Tan T, Chow ZL, et al. Multiplex immunohistochemistry/immunofluorescence (mIHC/IF) for PD-L1 testing in triple-negative breast cancer: a translational assay compared with conventional IHC. J Clin Pathol 2020;73:557–62. doi:10.1136/jclinpath-2019-206252.

Phillips D, Schürch CM, Khodadoust MS, et al. Highly multiplexed phenotyping of immunoregulatory proteins in the tumor microenvironment by CODEX tissue imaging. Front Immunol 2021;0. doi:10.3389/fimmu.2021.687673.

Ng HHM, Lee RY, Goh S, et al. Immunohistochemical scoring of CD38 in the tumor microenvironment predicts responsiveness to anti-PD-1/PD-L1 immunotherapy in hepatocellular carcinoma. J Immunother Cancer 2020;8:e000987. doi:10.1136/jitc-2020-000987

Abstract 627 Figure 1

Key target users and applications of ImmunoAtlas

Abstract 627 Figure 2

ImmunoAtlas' web interface for sharing and visualizing mIHC

Leveraging advances in immunopathology and artificial intelligence to analyze in vitro tumor models in composition and space

Cancer is the leading cause of death worldwide. Unfortunately, efforts to understand this disease are confounded by the complex, heterogenous tumor microenvironment (TME). Better understanding of the TME could lead to novel diagnostic, prognostic, and therapeutic discoveries. One way to achieve this involves in vitro tumor models that recapitulate the in vivo TME composition and spatial arrangement. Here, we review the potential of harnessing in vitro tumor models and artificial intelligence to delineate the TME. This includes (i) identification of novel features, (ii) investigation of higher-order relationships, and (iii) analysis and interpretation of multiomics data in a (iv) holistic, objective, reproducible, and efficient manner, which surpasses previous methods of TME analysis. We also discuss limitations of this approach, namely inadequate datasets, indeterminate biological correlations, ethical concerns, and logistical constraints; finally, we speculate on future avenues of research that could overcome these limitations, ultimately translating to improved clinical outcomes.

Utility of In Vitro Bioactivity as a Lower Bound Estimate of In Vivo Adverse Effect Levels and in Risk-Based Prioritization

Use of high-throughput, in vitro bioactivity data in setting a point-of-departure (POD) has the potential to accelerate the pace of human health safety evaluation by informing screening-level assessments. The primary objective of this work was to compare PODs based on high-throughput predictions of bioactivity, exposure predictions, and traditional hazard information for 448 chemicals. PODs derived from new approach methodologies (NAMs) were obtained for this comparison using the 50th (PODNAM, 50) and the 95th (PODNAM, 95) percentile credible interval estimates for the steady-state plasma concentration used in in vitro to in vivo extrapolation of administered equivalent doses. Of the 448 substances, 89% had a PODNAM, 95 that was less than the traditional POD (PODtraditional) value. For the 48 substances for which PODtraditional < PODNAM, 95, the PODNAM and PODtraditional were typically within a factor of 10 of each other, and there was an enrichment of chemical structural features associated with organophosphate and carbamate insecticides. When PODtraditional < PODNAM, 95, it did not appear to result from an enrichment of PODtraditional based on a particular study type (eg, developmental, reproductive, and chronic studies). Bioactivity:exposure ratios, useful for identification of substances with potential priority, demonstrated that high-throughput exposure predictions were greater than the PODNAM, 95 for 11 substances. When compared with threshold of toxicological concern (TTC) values, the PODNAM, 95 was greater than the corresponding TTC value 90% of the time. This work demonstrates the feasibility, and continuing challenges, of using in vitro bioactivity as a protective estimate of POD in screening-level assessments via a case study.

Virtual screening of potentially endocrine-disrupting chemicals against nuclear receptors and its application to identify PPARγ-bound fatty acids

Nuclear receptors (NRs) are key regulators of energy homeostasis, body development, and sexual reproduction. Xenobiotics binding to NRs may disrupt natural hormonal systems and induce undesired adverse effects in the body. However, many chemicals of concerns have limited or no experimental data on their potential or lack-of-potential endocrine-disrupting effects. Here, we propose a virtual screening method based on molecular docking for predicting potential endocrine-disrupting chemicals (EDCs) that bind to NRs. For 12 NRs, we systematically analyzed how multiple crystal structures can be used to distinguish actives and inactives found in previous high-throughput experiments. Our method is based on (i) consensus docking scores from multiple structures at a single functional state (agonist-bound or antagonist-bound), (ii) multiple functional states (agonist-bound and antagonist-bound), and (iii) multiple pockets (orthosteric site and alternative sites) of these NRs. We found that the consensus enrichment from multiple structures is better than or comparable to the best enrichment from a single structure. The discriminating power of this consensus strategy was further enhanced by a chemical similarity-weighted scoring scheme, yielding better or comparable enrichment for all studied NRs. Applying this optimized method, we screened 252 fatty acids against peroxisome proliferator-activated receptor gamma (PPARγ) and successfully identified 3 previously unknown fatty acids with Kd = 100-250 μM including two furan fatty acids: furannonanoic acid (FNA) and furanundecanoic acid (FUA), and one cyclopropane fatty acid: phytomonic acid (PTA). These results suggested that the proposed method can be used to rapidly screen and prioritize potential EDCs for further experimental evaluations.

PI3K Catalytic Subunits α and β Modulate Cell Death and IL-6 Secretion Induced by Talc Particles in Human Lung Carcinoma Cells

Hydrated magnesium silicate (or "talc" particles) is a sclerosis agent commonly used in the management of malignant pleural effusions, a common symptom of metastatic diseases, including lung cancers. However, the direct effects of talc particles to lung carcinoma cells, which can be found in the malignant pleural effusion fluids from patients with lung cancer, are not fully understood. Here, we report a study of the signaling pathways that can modulate the cell death and IL-6 secretion induced by talc particles in human lung carcinoma cells. We found that talc-sensitive cells have higher mRNA and protein expression of PI3K catalytic subunits α and β. Further experiments confirmed that modulation (inhibition or activation) of the PI3K pathway reduces or enhances cellular sensitivity to talc particles, respectively, independent of the inflammasome. By knocking down specific PI3K isoforms, we also confirmed that both PI3Kα and -β mediate the observed talc effects. Our results suggest a novel role of the PI3K pathway in talc-induced cell death and IL-6 secretion in lung carcinoma cells. These cellular events are known to drive fibrosis, and thus further studies of the PI3K pathway may provide a better understanding of the mechanisms of talc sclerosis in the malignant pleural space.

Predicting direct hepatocyte toxicity in humans by combining high-throughput imaging of HepaRG cells and machine learning-based phenotypic profiling

Accurate prediction of drug- and chemical-induced hepatotoxicity remains to be a problem for pharmaceutical companies as well as other industries and regulators. The goal of the current study was to develop an in vitro/in silico method for the rapid and accurate prediction of drug- and chemical-induced hepatocyte injury in humans. HepaRG cells were employed for high-throughput imaging in combination with phenotypic profiling. A reference set of 69 drugs and chemicals was screened at a range of 7 concentrations, and the cellular response values were used for training a supervised classifier and for determining assay performance by using tenfold cross-validation. The results showed that the best performing phenotypic features were related to nuclear translocation of RELA (RELA proto-oncogene, NF-kB subunit; also known as NF-kappa B p65), DNA organization, and the F-actin cytoskeleton. Using a subset of 30 phenotypic features, direct hepatocyte toxicity in humans could be predicted with a test sensitivity, specificity and balanced accuracy of 73%, 92%, and 83%, respectively. The method was applied to another set of 26 drugs and chemicals with unclear annotation and their hepatocyte toxicity in humans was predicted. The results also revealed that the identified discriminative phenotypic changes were related to cell death and cellular senescence. Whereas cell death-related endpoints are widely applied in in vitro toxicology, cellular senescence-related endpoints are not, although cellular senescence can be induced by various drugs and other small molecule compounds and plays an important role in liver injury and disease. These findings show how phenotypic profiling can reveal unexpected chemical-induced mechanisms in toxicology.

A Case Study with Triazole Fungicides to Explore Practical Application of Next-Generation Hazard Assessment Methods for Human Health

The ongoing developments in chemical risk assessment have led to new concepts building on integration of sophisticated nonanimal models for hazard characterization. Here we explore a pragmatic approach for implementing such concepts, using a case study of three triazole fungicides, namely, flusilazole, propiconazole, and cyproconazole. The strategy applied starts with evaluating the overall level of concern by comparing exposure estimates to toxicological potential, followed by a combination of in silico tools and literature-derived high-throughput screening assays and computational elaborations to obtain insight into potential toxicological mechanisms and targets in the organism. Additionally, some targeted in vitro tests were evaluated for their utility to confirm suspected mechanisms of toxicity and to generate points of departure. Toxicological mechanisms instead of the current "end point-by-end point" approach should guide the selection of methods and assays that constitute a toolbox for next-generation risk assessment. Comparison of the obtained in silico and in vitro results with data from traditional in vivo testing revealed that, overall, nonanimal methods for hazard identification can produce adequate qualitative hazard information for risk assessment. Follow-up studies are needed to further refine the proposed approach, including the composition of the toolbox, toxicokinetics models, and models for exposure assessment.

Building predictive in vitro pulmonary toxicity assays using high-throughput imaging and artificial intelligence

Human lungs are susceptible to the toxicity induced by soluble xenobiotics. However, the direct cellular effects of many pulmonotoxic chemicals are not always clear, and thus, a general in vitro assay for testing pulmonotoxicity applicable to a wide variety of chemicals is not currently available. Here, we report a study that uses high-throughput imaging and artificial intelligence to build an in vitro pulmonotoxicity assay by automatically comparing and selecting human lung-cell lines and their associated quantitative phenotypic features most predictive of in vivo pulmonotoxicity. This approach is called “High-throughput In vitro Phenotypic Profiling for Toxicity Prediction” (HIPPTox). We found that the resulting assay based on two phenotypic features of a human bronchial epithelial cell line, BEAS-2B, can accurately classify 33 reference chemicals with human pulmonotoxicity information (88.8% balance accuracy, 84.6% sensitivity, and 93.0% specificity). In comparison, the predictivity of a standard cell-viability assay on the same set of chemicals is much lower (77.1% balanced accuracy, 84.6% sensitivity, and 69.5% specificity). We also used the assay to evaluate 17 additional test chemicals with unknown/unclear human pulmonotoxicity, and experimentally confirmed that many of the pulmonotoxic reference and predicted-positive test chemicals induce DNA strand breaks and/or activation of the DNA-damage response (DDR) pathway. Therefore, HIPPTox helps us to uncover these common modes-of-action of pulmonotoxic chemicals. HIPPTox may also be applied to other cell types or models, and accelerate the development of predictive in vitro assays for other cell-type- or organ-specific toxicities.

High-throughput prediction of nephrotoxicity in humans

The Lush Science Prize 2016 was awarded to Daniele Zink and Lit-Hsin Loo for the interdisciplinary and collaborative work between their research groups in developing alternative methods for the prediction of nephrotoxicity in humans. The collaboration has led to the establishment of a series of pioneering alternative methods for nephrotoxicity prediction, which includes: predictive gene expression markers based on pro-inflammatory responses; predictive in vitro cellular models based on pluripotent stem cell-derived proximal tubular-like cells; and predictive cellular phenotypic markers based on chromatin and cytoskeletal changes. A high-throughput method was established for chemical testing, which is currently being used to predict the potential human nephrotoxicity of ToxCast compounds in collaboration with the US Environmental Protection Agency. Similar high-throughput imaging-based methodologies are currently being developed and adapted by the Zink and Loo groups, to include other human organs and cell types. The ultimate goal is to develop a portfolio of methods accepted for the accurate prediction of human organ-specific toxicity and the consequent replacement of animal experiments.

Early spatiotemporal-specific changes in intermediate signals are predictive of cytotoxic sensitivity to TNFα and co-treatments

Signaling pathways can generate different cellular responses to the same cytotoxic agents. Current quantitative models for predicting these differential responses are usually based on large numbers of intracellular gene products or signals at different levels of signaling cascades. Here, we report a study to predict cellular sensitivity to tumor necrosis factor alpha (TNFα) using high-throughput cellular imaging and machine-learning methods. We measured and compared 1170 protein phosphorylation events in a panel of human lung cancer cell lines based on different signals, subcellular regions, and time points within one hour of TNFα treatment. We found that two spatiotemporal-specific changes in an intermediate signaling protein, p90 ribosomal S6 kinase (RSK), are sufficient to predict the TNFα sensitivity of these cell lines. Our models could also predict the combined effects of TNFα and other kinase inhibitors, many of which are not known to target RSK directly. Therefore, early spatiotemporal-specific changes in intermediate signals are sufficient to represent the complex cellular responses to these perturbations. Our study provides a general framework for the development of rapid, signaling-based cytotoxicity screens that may be used to predict cellular sensitivity to a cytotoxic agent, or identify co-treatments that may sensitize or desensitize cells to the agent.

Large-scale image-based screening and profiling of cellular phenotypes

Cellular phenotypes are observable characteristics of cells resulting from the interactions of intrinsic and extrinsic chemical or biochemical factors. Image-based phenotypic screens under large numbers of basal or perturbed conditions can be used to study the influences of these factors on cellular phenotypes. Hundreds to thousands of phenotypic descriptors can also be quantified from the images of cells under each of these experimental conditions. Therefore, huge amounts of data can be generated, and the analysis of these data has become a major bottleneck in large-scale phenotypic screens. Here, we review current experimental and computational methods for large-scale image-based phenotypic screens. Our focus is on phenotypic profiling, a computational procedure for constructing quantitative and compact representations of cellular phenotypes based on the images collected in these screens. © 2016 International Society for Advancement of Cytometry.

Supervised prediction of drug-induced nephrotoxicity based on interleukin-6 and -8 expression levels

Background

Drug-induced nephrotoxicity causes acute kidney injury and chronic kidney diseases, and is a major reason for late-stage failures in the clinical trials of new drugs. Therefore, early, pre-clinical prediction of nephrotoxicity could help to prioritize drug candidates for further evaluations, and increase the success rates of clinical trials. Recently, an in vitro model for predicting renal-proximal-tubular-cell (PTC) toxicity based on the expression levels of two inflammatory markers, interleukin (IL)-6 and -8, has been described. However, this and other existing models usually use linear and manually determined thresholds to predict nephrotoxicity. Automated machine learning algorithms may improve these models, and produce more accurate and unbiased predictions.

Results

Here, we report a systematic comparison of the performances of four supervised classifiers, namely random forest, support vector machine, k-nearest-neighbor and naive Bayes classifiers, in predicting PTC toxicity based on IL-6 and -8 expression levels. Using a dataset of human primary PTCs treated with 41 well-characterized compounds that are toxic or not toxic to PTC, we found that random forest classifiers have the highest cross-validated classification performance (mean balanced accuracy = 87.8%, sensitivity = 89.4%, and specificity = 85.9%). Furthermore, we also found that IL-8 is more predictive than IL-6, but a combination of both markers gives higher classification accuracy. Finally, we also show that random forest classifiers trained automatically on the whole dataset have higher mean balanced accuracy than a previous threshold-based classifier constructed for the same dataset (99.3% vs. 80.7%).

Conclusions

Our results suggest that a random forest classifier can be used to automatically predict drug-induced PTC toxicity based on the expression levels of IL-6 and -8.

Quantitative protein localization signatures reveal an association between spatial and functional divergences of proteins

Protein subcellular localization is a major determinant of protein function. However, this important protein feature is often described in terms of discrete and qualitative categories of subcellular compartments, and therefore it has limited applications in quantitative protein function analyses. Here, we present Protein Localization Analysis and Search Tools (PLAST), an automated analysis framework for constructing and comparing quantitative signatures of protein subcellular localization patterns based on microscopy images. PLAST produces human-interpretable protein localization maps that quantitatively describe the similarities in the localization patterns of proteins and major subcellular compartments, without requiring manual assignment or supervised learning of these compartments. Using the budding yeast Saccharomyces cerevisiae as a model system, we show that PLAST is more accurate than existing, qualitative protein localization annotations in identifying known co-localized proteins. Furthermore, we demonstrate that PLAST can reveal protein localization-function relationships that are not obvious from these annotations. First, we identified proteins that have similar localization patterns and participate in closely-related biological processes, but do not necessarily form stable complexes with each other or localize at the same organelles. Second, we found an association between spatial and functional divergences of proteins during evolution. Surprisingly, as proteins with common ancestors evolve, they tend to develop more diverged subcellular localization patterns, but still occupy similar numbers of compartments. This suggests that divergence of protein localization might be more frequently due to the development of more specific localization patterns over ancestral compartments than the occupation of new compartments. PLAST enables systematic and quantitative analyses of protein localization-function relationships, and will be useful to elucidate protein functions and how these functions were acquired in cells from different organisms or species. A public web interface of PLAST is available at http://plast.bii.a-star.edu.sg.

Proteins are fundamental building blocks of cells. They perform a variety of biological functions, many of which are essential to the vitality and normal functioning of cells. Proteins have to be located at the appropriate regions inside a cell to perform their functions. Therefore, when proteins change their locations, they may acquire new or different functions. However, the relationships between the locations and functions of proteins are difficult to analyze because protein locations are often represented in distinct and manually-defined categories of subcellular regions. Many proteins have complex or subtle differences in their localization patterns that can be hardly represented by these categories. Here, we present an automated analysis tool for generating quantitative signatures of protein localization patterns without requiring manual or automated assignments of proteins into distinct categories. We show that our tool can identify proteins located at the same subcellular regions more accurately than existing categorization-based methods. Our tool allows comprehensive and more accurate studies of the relationships between protein localizations and functions. By knowing where proteins are located and how their locations were changed, we may discover their functions and better understand how they acquire these functions.

cellXpress: a fast and user-friendly software platform for profiling cellular phenotypes

BACKGROUND

High-throughput, image-based screens of cellular responses to genetic or chemical perturbations generate huge numbers of cell images. Automated analysis is required to quantify and compare the effects of these perturbations. However, few of the current freely-available bioimage analysis software tools are optimized for efficient handling of these images. Even fewer of them are designed to transform the phenotypic features measured from these images into discriminative profiles that can reveal biologically meaningful associations among the tested perturbations.

RESULTS

We present a fast and user-friendly software platform called "cellXpress" to segment cells, measure quantitative features of cellular phenotypes, construct discriminative profiles, and visualize the resulting cell masks and feature values. We have also developed a suite of library functions to load the extracted features for further customizable analysis and visualization under the R computing environment. We systematically compared the processing speed, cell segmentation accuracy, and phenotypic-profile clustering performance of cellXpress to other existing bioimage analysis software packages or algorithms. We found that cellXpress outperforms these existing tools on three different bioimage datasets. We estimate that cellXpress could finish processing a genome-wide gene knockdown image dataset in less than a day on a modern personal desktop computer.

CONCLUSIONS

The cellXpress platform is designed to make fast and efficient high-throughput phenotypic profiling more accessible to the wider biological research community. The cellXpress installation packages for 64-bit Windows and Linux, user manual, installation guide, and datasets used in this analysis can be downloaded freely from http://www.cellXpress.org.

Heterogeneity in the physiological states and pharmacological responses of differentiating 3T3-L1 preadipocytes

A systems biology–based analysis shows that differentiating adipocytes look very different at the single-cell level and form distinct cellular subpopulations.

Increases in key components of adipogenesis and lipolysis pathways correlate at the population-averaged level during adipogenesis. However, differentiating preadipocytes are highly heterogeneous in cellular and lipid droplet (LD) morphologies, and the degree to which individual cells follow population-averaged trends is unclear. In this study, we analyze the molecular heterogeneity of differentiating 3T3-L1 preadipocytes using immunofluorescence microscopy. Unexpectedly, we only observe a small percentage of cells with high simultaneous expression of markers for adipogenesis (peroxisome proliferator-activated receptor γ [PPARγ], CCAAT/enhancer-binding protein α, and adiponectin) and lipid accumulation (hormone-sensitive lipase, perilipin A, and LDs). Instead, we identify subpopulations of cells with negatively correlated expressions of these readouts. Acute perturbation of adipocyte differentiation with PPARγ agonists, forskolin, and fatty acids induced subpopulation-specific effects, including redistribution of the percentage of cells in observed subpopulations and differential expression levels of PPARγ. Collectively, our results suggested that heterogeneity observed during 3T3-L1 adipogenesis reflects a dynamic mixture of subpopulations with distinct physiological states.

An approach for extensibly profiling the molecular states of cellular subpopulations

Microscopy often reveals the existence of phenotypically distinct cellular subpopulations. However, additional characterization of observed subpopulations can be limited by the number of biomolecular markers that can be simultaneously monitored. Here we present a computational approach for extensibly profiling cellular subpopulations by freeing one or more imaging channels to monitor additional probes. In our approach, we trained classifiers to re-identify subpopulations accurately based on an enhanced collection of phenotypic features extracted from only a subset of the original markers. Then we constructed subpopulation profiles step-wise from replicate experiments, in which cells were labeled with different but overlapping marker sets. We applied our approach to identify molecular differences among subpopulations and to identify functional groupings of markers, in populations of differentiating mouse preadipocytes, polarizing human neutrophil-like cells and dividing human cancer cells.

Image-based multivariate profiling of drug responses from single cells

Quantitative analytical approaches for discovering new compound mechanisms are required for summarizing high-throughput, image-based drug screening data. Here we present a multivariate method for classifying untreated and treated human cancer cells based on approximately 300 single-cell phenotypic measurements. This classification provides a score, measuring the magnitude of the drug effect, and a vector, indicating the simultaneous phenotypic changes induced by the drug. These two quantities were used to characterize compound activities and identify dose-dependent multiphasic responses. A systematic survey of profiles extracted from a 100-compound compendium of image data revealed that only 10-15% of the original features were required to detect a compound effect. We report the most informative image features for each compound and fluorescence marker set using a method that will be useful for determining minimal collections of readouts for drug screens. Our approach provides human-interpretable profiles and automatic determination of on- and off-target effects.

New criteria for selecting differentially expressed genes

One of the major concerns in detecting changes in higher moments is these changes may be due to outliers or process errors that are not biologically significant. For example, a larger variance observed in the expression levels may simply due to the larger variation in the data collecting process. Several outliers, which exhibit some extreme expression levels than the rest of the samples, may also increase the variance or skewness of the expression levels significantly. So it is very important to reduce the effect of outliers and process errors by proper experimental designs [27], such as technical replicates and biological replicates, before high sensitivity criterion, such as ADS, can be applied. We have presented and demonstrated the operation of two new criteria, ADS and the MDS, for identifying differentially expressed genes. These two criteria were compared with several commonly used criteria, namely WTS, WRS, FCS, and ICE. Experiments with simulated data show ADS to be more powerful than the WTS. When high-sensitivity screening is required, ADS appears to be preferable to WTS. When an FPR similar to WTS is desired, MDS should be used. The popular Wilcoxon rank sum is a more conservative approach that should be employed when the lowest FPR is desired, even at the expense of lower TPRs. ICE is a less desirable criterion because it does not perform well for data generated by the normal model. FCS gave results similar to those of WTS. Evaluation of these algorithms using real biological datasets showed that ADS and MDS flagged several biologically significant genes that were missed by WTS, besides selecting most of the genes that are also selected by WTS.

Quality of life in cadaver and living-related renal transplant recipients in Kuala Lumpur hospital

OBJECTIVE

To examine the quality of life in cadaver (CAD) and living-related (LRRT) renal transplant recipients.

METHODS

A cross-sectional study was done on patients followed in renal transplant clinic from 1/4/03 to 1/7/03 using the SF-36 questionnaire. Inclusion criteria were age >16 years,minimum of 3 months' posttransplant, and informed consent. Exclusion criteria were current treatment for rejection or infection or any life-threatening conditions. Information on duration of transplant, duration of dialysis prior to transplant, number of co-morbidities, and sociodemodraphic data were collected.

RESULTS

Sixty-four among 110 patients (58.1%) completed the SF36 questionnaire. The LRRT recipients were younger, had a longer duration of transplant, and had spent significantly less time on dialysis prior to transplant compared to CAD transplant patients. Overall, the physical composite and the mental composite scores were not significantly different between the two transplant groups. Age was negatively associated with the physical composite score (Spearman's rho -0.251, P < .05) and bodily pain (Spearman's rho -0.266, P < .05). Duration of dialysis prior to transplant was negatively correlated with social functioning (Spearman's rho -0.28, P < .05) and mental health (Spearman's rho -0.39, P < .005). In multiple regression analysis, age was a significant predictor of the SF36 physical composite score (P < .05).

CONCLUSION

This study shows that the quality of life between LRRT and CAD recipients was not significantly different. Increased age was associated with poorer physical capacity.

Infection of central nervous system by motile Enterococcus: first case report

A 66-year-old man with four indwelling ventriculoperitoneal shunts for multiloculated hydrocephalus from a complicated case of meningitis a year before developed shunt infection based on a syndrome of fever, drowsiness, and cerebrospinal fluid neutrophil pleocytosis in the background of repeated surgical manipulation to relieve successive shunt blockages. The cerebrospinal fluid culture, which yielded a motile Enterococcus species, was believed to originate from the gut. This isolate was lost in storage and could not be characterized further. The patient improved with vancomycin and high-dose ampicillin therapy. He relapsed a month later with Enterococcus gallinarum shunt infection, which responded to high-dose ampicillin and gentamicin therapy. This is probably the first case report of motile Enterococcus infection of the central nervous system.

Phylogenetic relationships of Salmonella typhi and Salmonella typhimurium based on 16S rRNA sequence analysis

The 16S rRNA gene sequences of Salmonella typhi and Salmonella typhimurium were amplified by PCR, cloned, and sequenced. These sequences were analyzed by comparison with reference organisms from the family Enterobacteriaceae. Both S. typhi and S. typhimurium belong to the gamma subdivision of the class Proteobacteria.

Comparison of multiplex PCR and culture for detection of Legionellae in cooling tower water samples

We compared multiplex polymerase chain reaction (PCR) and culture for detecting the presence of Legionella pneumophila and Legionella spp in cooling tower water samples. Multiplex PCR was performed after phenol extraction of DNA from the samples. The set of primers for the PCR assay involved the 5S rRNA (Legionella spp) and the mip (macrophage infectivity potentiator gene, specific for L. pneumophila) genes as target sequences for amplification. Both the sensitivity and the specificity of the PCR assay were 100% when the 5S rRNA gene was used as target sequence. Isolation of Legionellae from the samples was observed only with the PCR-positive samples. We propose that PCR be used as a screening test before attempting to culture Legionellae from cooling tower water samples.