Colin Craig McCulloch

Highly multiplexed single-cell analysis of formalin-fixed, paraffin-embedded cancer tissue

Limitations on the number of unique protein and DNA molecules that can be characterized microscopically in a single tissue specimen impede advances in understanding the biological basis of health and disease. Here we present a multiplexed fluorescence microscopy method (MxIF) for quantitative, single-cell, and subcellular characterization of multiple analytes in formalin-fixed paraffin-embedded tissue. Chemical inactivation of fluorescent dyes after each image acquisition round allows reuse of common dyes in iterative staining and imaging cycles. The mild inactivation chemistry is compatible with total and phosphoprotein detection, as well as DNA FISH. Accurate computational registration of sequential images is achieved by aligning nuclear counterstain-derived fiducial points. Individual cells, plasma membrane, cytoplasm, nucleus, tumor, and stromal regions are segmented to achieve cellular and subcellular quantification of multiplexed targets. In a comparison of pathologist scoring of diaminobenzidine staining of serial sections and automated MxIF scoring of a single section, human epidermal growth factor receptor 2, estrogen receptor, p53, and androgen receptor staining by diaminobenzidine and MxIF methods yielded similar results. Single-cell staining patterns of 61 protein antigens by MxIF in 747 colorectal cancer subjects reveals extensive tumor heterogeneity, and cluster analysis of divergent signaling through ERK1/2, S6 kinase 1, and 4E binding protein 1 provides insights into the spatial organization of mechanistic target of rapamycin and MAPK signal transduction. Our results suggest MxIF should be broadly applicable to problems in the fields of basic biological research, drug discovery and development, and clinical diagnostics.

Genetic versus chemoprotective activation of Nrf2 signaling: overlapping yet distinct gene expression profiles between Keap1 knockout and triterpenoid-treated mice

Loss of NF-E2-related factor 2 (Nrf2) signaling increases susceptibility to acute toxicity, inflammation and carcinogenesis in mice due to the inability to mount adaptive responses. In contrast, disruption of Keap1 (a cytoplasmic modifier of Nrf2 turnover) protects against these stresses in mice, although inactivating mutations in Keap1 have been identified recently in some human cancers. Global characterization of Nrf2 activation is important to exploit this pathway for chemoprevention in healthy, yet at-risk individuals and also to elucidate the consequences of hijacking the pathway in Keap1-mutant human cancers. Liver-targeted conditional Keap1-null, Albumin-Cre:Keap1((flox/-)) (CKO) mice provide a model of genetic activation of Nrf2 signaling. By coupling global gene expression analysis of CKO mice with analysis of pharmacologic activation using the synthetic oleanane triterpenoid 1-[2-cyano-3,12-dioxooleana-1,9(11)-dien-28-oyl]imidazole (CDDO-Im), we are able to gain insight into pathways affected by Nrf2 activation. CDDO-Im is an extremely potent activator of Nrf2 signaling. CKO mice were used to identify genes modulated by genetic activation of Nrf2 signaling. The CKO response was compared with hepatic global gene expression changes in wild-type mice treated with CDDO-Im at a maximal Nrf2 activating dose. The results show that genetic and pharmacologic activation of Nrf2 signaling modulates pathways beyond detoxication and cytoprotection, with the largest cluster of genes associated with lipid metabolism. Genetic activation of Nrf2 results in much larger numbers of detoxication and lipid metabolism gene changes. Additionally, analysis of pharmacologic activation suggests that Nrf2 is the primary mediator of CDDO-Im activity, though other cell-signaling targets are also modulated following an oral dose of 30 micromol/kg.

Model-based detection of lung nodules in computed tomography exams1

Abstract Rationale and objectives In this study, we developed a prototype model-based computer aided detection (CAD) system designed to automatically detect both solid and subsolid pulmonary nodules in computed tomography (CT) images. By using this CAD algorithm, along with the radiologist’s initial interpretation, we aim to improve the sensitivity of radiologic readings of CT lung exams. Materials and methods We have developed a model-based CAD algorithm through the use of precise mathematic models that capture scanner physics and anatomic information. Our model-based CAD algorithm uses multiple segmentation algorithms to extract noteworthy structures in the lungs and a Bayesian statistical model selection framework to determine the probability of various anatomical events throughout the lung. We tested this algorithm on 50 low-dose CT lung cancer screening cases in which ground truth was produced through readings by three expert chest radiologists. Results Using this model-based CAD algorithm on 50 low-dose CT cases, we measured potential sensitivity improvements of 7% and 5% in two radiologists with respect to all noncalcified nodules, solid and subsolid, greater than 5 mm in diameter. The third radiologist did not miss any nodules in the ground truth set. The CAD algorithm produced 8.3 false positives per case. Conclusion Our prototype CAD system demonstrates promising results as a tool to improve the quality of radiologic readings by increasing radiologist sensitivity. A significant advantage of this model-based approach is that it can be easily extended to support additional anatomic models as clinical understanding and scanning practices improve.

Divergent Transcriptomic Responses to Aryl Hydrocarbon Receptor Agonists Between Rat and Human Primary Hepatocytes

Toxicogenomics has great potential for enhancing our understanding of environmental chemical toxicity, hopefully leading to better informed human health risk assessments. This study employed toxicogenomic technology to reveal species differences in response to two prototypical aryl hydrocarbon receptor (AHR) agonists 2,3,7,8-tetrachlorodibenzo-p-dioxin and the polychlorinated biphenyl (PCB) congener PCB 126. Dose-responses of primary cultures of rat and human hepatocytes were determined using species-specific microarrays sharing over 4000 gene orthologs. Forty-seven human and 79 rat genes satisfied dose-response criteria for both chemicals and were subjected to further analysis including the calculation of the 50% effective concentration and the relative potency (REP) of PCB 126 for each gene. Only five responsive orthologous genes were shared between the two species; yet, the geometric mean of the REPs for all rat and human modeled responsive genes were 0.06 (95% confidence interval [CI]; 0.03-0.1) and 0.002 (95% CI; 0.001-0.005), respectively, suggesting broad species differences in the initial events that follow AHR activation but precede toxicity. This indicates that there are species differences in both the specific genes that responded and the agonist potency and REP for those genes. This observed insensitivity of human cells to PCB 126 is consistent with more traditional measurements of AHR activation (i.e., cytochrome P450 1A1 enzyme activity) and suggests that the species difference in PCB 126 sensitivity is likely due to certain aspects of AHR function. That a species divergence also exists in this expanded AHR-regulated gene repertoire is a novel finding and should help when extrapolating animal data to humans.

Toxicogenomic Analysis of Gender, Chemical, and Dose Effects in Livers of TCDD- or Aroclor 1254–Exposed Rats Using a Multifactor Linear Model

Chronic exposure of Sprague-Dawley (SD) rats to either 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) or Aroclor 1254 results in female-selective induction of hepatic tumors. The relative potency of dioxins and polychlorinated biphenyl mixtures, such as Aroclor 1254, is often estimated using the internationally endorsed toxic equivalency (TEQ) approach. Comparing the genome wide changes in gene expression in both genders following exposure to TEQ doses of these chemicals should identify critical sets of early response genes while further defining the concept of the TEQ of halogenated aromatic hydrocarbons. Aroclor 1254 at 0.6, 6.0, and 60 mg/kg body weight and TEQ doses of TCDD (0.3 and 3.0 mug/kg), calculated to match the top two Aroclor 1254 doses, were orally administered to SD rats for three consecutive days. Day 4 gene expression in hepatic tissue was determined using microarrays. A linear mixed-effects statistical model was developed to analyze the data in relation to treatment, gender, and gender * treatment (G*T) interactions. The genes most changed included 54 genes with and 51 genes without a significant model G*T term. The known aryl hydrocarbon receptor (AHR) battery genes (Cyp1a1, Cyp1a2, Cyp1b1, Aldh3a1), and novel genes, responded in a TEQ dose-dependent manner in both genders. However, an important observation was the apparent disruption of sexually dimorphic basal gene expression, particularly for female rats. Because many of these genes are involved in steroid metabolism, exposure to either TCDD or Aroclor 1254 could disrupt proliferative signals more in female rats as a possible consequence of altered estrogen metabolism. This study extends the findings of previous rodent bioassays by identifying groups of genes, other than the well-characterized AHR response genes, whose disruption may be important in the tumorigenic mechanism in this rat strain.

One step closer to fixing association studies: evidence for age- and gender-specific allele frequency variations and deviations from Hardy-Weinberg expectations in controls

Association studies are the most powerful method available for identifying modest gene effects in complex disorders, but they often produce inconsistent results. With the rapidly growing SNP databases, haplotype maps and high throughput genotyping, the use of association studies is expected to increase; therefore, it is critical and timely that the problems with study design are identified and fixed. We questioned if unrecognized allele and genotype frequency variations in controls could be responsible for some of the inconsistent association findings. We performed a population genetic study of apolipoprotein E (APOE) and cytochrome P450 2D6 (CYP2D6) in 1,748 individuals ranging in age from newborns to centenarians. Although APOE and CYP2D6 are two of the most commonly used candidate genes, this is the first study to examine age- and gender-specific frequency distributions over the entire age spectrum, using a large, ethnically and geographically uniform population. We found significant, previously unrecognized variations in APOE allele frequencies, and deviations from Hardy-Weinberg expectations in CYP2D6 genotype frequencies starting at birth. The allele frequency variations within controls were larger than some reported case-control differences. We demonstrate that unrecognized frequency fluctuations in controls are a serious and potentially common confounder whose impact on association studies has not been appreciated, and one that can be addressed with proper study design. We recommend that population genetic studies be performed on commonly used candidate markers and that rigorous standards be applied for case-control matching.

Visualizing Disease Associations: Graphic Analysis of Frequency Distributions as a Function of Age Using Moving Average Plots (MAP) with Application to Alzheimer's and Parkinson's Disease

Age‐related variation in marker frequency can be a confounder in association studies, leading to both false‐positive and false‐negative findings and subsequently to inconsistent reproducibility. We have developed a simple method, based on a novel extension of moving average plots (MAP), which allows investigators to inspect the frequency data for hidden age‐related variations. MAP uses the standard case‐control association data and generates a birds‐eye view of the frequency distributions across the age spectrum; a picture in which one can see if, how, and when the marker frequencies in cases differ from that in controls. The marker can be specified as an allele, genotype, haplotype, or environmental factor; and age can be age‐at‐onset, age when subject was last known to be unaffected, or duration of exposure. Signature patterns that emerge can help distinguish true disease associations from spurious associations due to age effects, age‐varying associations from associations that are uniform across all ages, and associations with risk from associations with age‐at‐onset. Utility of MAP is illustrated by application to genetic and epidemiological association data for Alzheimers and Parkinsons disease. MAP is intended as a descriptive method, to complement standard statistical techniques. Although originally developed for age patterns, MAP is equally useful for visualizing any quantitative trait. Genet. Epidemiol. 34:92–99, 2010.

Towards wind farm performance optimization through empirical models

Wind Turbine performance improvement measurements are challenging, especially when improvements affect air flow to the nacelle anemometer sensor which is often used to baseline performance. Uncertainty in this area can impede optimization of wind farms by making it difficult to show the benefit of upgrades to individual turbines, jointly optimize wind turbine performance in a farm, and validate the effects of optimization algorithms - particularly farm level algorithms and strategies that mitigate waking affects. In this paper we introduce methods that augment traditional methods for baselining wind turbine performance using multi-feature estimation based on empirical data and present a method for normalizing AEP uncertainty estimates. This innovative method does not rely solely on nacelle anemometer estimates or expensive additional sensors, as has been the historical approach but can leverage these trusted sensors if they are available. Future directions for whole farm optimizations are discussed.

Model-based detection of lung lesions in CT exams

Abstract The thorough detection of nodules in high-resolution CT lung scans is an increasingly difficult, labor-intensive, and critical radiological task. Recent clinical research on early lung cancer CT presentation has demonstrated the significant clinical need to detect the more subtle subsolid nodules as well as the traditional solid nodule. We have developed a model-based computer-aided detection (CAD) algorithm designed to automatically detect both of these nodule presentation types through the use of precise mathematical models that capture scanner physics and anatomy and pathology domain knowledge. Our model-based CAD algorithm utilizes a Bayesian framework for determining the probability of multiple competing anatomical and pathological events throughout the lung. Using this model-based CAD algorithm on 50 low-dose CT lung cancer screening cases, we measured a 3.9% average improvement in radiologist sensitivity (93.8% to 97.7%) with 8.3 false positives per case for all nodules ≥5 mm in size. This model-based approach can be easily extended to support additional anatomy and pathology models as clinical understanding and scanning practices improve.