Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to main content

    Jiang Gui

    Several new statistical methods have been developed to identify the overall impact of an exposure mixture on health outcomes. Weighted quantile sum (WQS) regression assigns the joint mixture effect weights to indicate the overall... more
    Several new statistical methods have been developed to identify the overall impact of an exposure mixture on health outcomes. Weighted quantile sum (WQS) regression assigns the joint mixture effect weights to indicate the overall association of multiple exposures, and quantile-based g-computation is a generalized version of WQS without the restriction of directional homogeneity. This paper proposes an adaptive-mixture-categorization (AMC)-based g-computation approach that combines g-computation with an optimal exposure categorization search using the F statistic. AMC-based g-computation reduces variance within each category and retains the variance between categories to build more powerful predictors. In a simulation study, the performance of association analysis was improved using categorizing by AMC compared with quantiles. We applied this method to assess the association between a mixture of 12 trace element concentrations measured from toenails and the risk of non-muscle invasiv...
    Genome-wide association studies (GWAS) identified the chromosome 15q25.1 locus as a leading susceptibility region for lung cancer. However, the pathogenic pathways, through which susceptibility SNPs within chromosome 15q25.1 affects lung... more
    Genome-wide association studies (GWAS) identified the chromosome 15q25.1 locus as a leading susceptibility region for lung cancer. However, the pathogenic pathways, through which susceptibility SNPs within chromosome 15q25.1 affects lung cancer risk, have not been explored. We analyzed three cohorts with GWAS data consisting 42,901 individuals and lung expression quantitative trait loci (eQTL) data on 409 individuals to identify and validate the underlying pathways and to investigate the combined effect of genes from the identified susceptibility pathways. The KEGG neuroactive ligand receptor interaction pathway, two Reactome pathways, and 22 Gene Ontology terms were identified and replicated to be significantly associated with lung cancer risk, with P values less than 0.05 and FDR less than 0.1. Functional annotation of eQTL analysis results showed that the neuroactive ligand receptor interaction pathway and gated channel activity were involved in lung cancer risk. These pathways p...
    The mTORC1 inhibitor RAD001 (everolimus) is approved for treatment of recurrent/metastatic estrogen receptor (ER)-positive breast cancer in combination with the aromatase inhibitor (AI) exemestane. The benefits of A) continued... more
    The mTORC1 inhibitor RAD001 (everolimus) is approved for treatment of recurrent/metastatic estrogen receptor (ER)-positive breast cancer in combination with the aromatase inhibitor (AI) exemestane. The benefits of A) continued anti-estrogen therapy for anti-estrogen-resistant disease in the context of mTORC1 inhibition, and B) adjuvant everolimus in combination with anti-estrogen therapy for early-stage disease are being tested clinically, but molecular rationale remains unclear. We hypothesized that mTORC1 inhibition activates the IGF-1R/InsR/IRS-1/2 axis in an ER-dependent manner to drive PI3K/AKT and promote cancer cell survival, implicating ER in survival signaling induced by mTORC1 inhibition. Anti-estrogen treatment synergized with RAD001 to inhibit ER+ breast cancer cell growth. Inhibition of ER, IGF-1R/InsR, or IRS-1/2 suppressed AKT activation induced by mTORC1 inhibition. RAD001 primed IGF-1R/InsR for activation, which was enhanced by ER signaling. Post-menopausal patients...
    While dynamic contrast-enhanced magnetic resonance imaging (DCE MRI) is recognized as the most sensitive examination for breast cancer detection, it has a substantial false positive rate and gadolinium (Gd) contrast agents are not... more
    While dynamic contrast-enhanced magnetic resonance imaging (DCE MRI) is recognized as the most sensitive examination for breast cancer detection, it has a substantial false positive rate and gadolinium (Gd) contrast agents are not universally well tolerated. As a result, alternatives to diagnosing breast cancer based on endogenous contrast are of growing interest. In this study, endogenous near-infrared spectral tomography (NIRST) guided by T2 MRI was evaluated to explore whether the combined imaging modality, which does not require contrast injection or involve ionizing radiation, can achieve acceptable diagnostic performance. Twenty-four subjects-16 with pathologically confirmed malignancy and 8 with benign abnormalities-were simultaneously imaged with MRI and NIRST prior to definitive pathological diagnosis. MRIs were evaluated independently by three breast radiologists blinded to the pathological results. Optical image reconstructions were constrained by grayscale values in the ...
    Polymorphisms in microRNAs and their target sites can disrupt microRNA-dependent gene regulation, and have been associated with cancer susceptibility. However, genome-scale analyses of microRNA-related genetic variation in cancer are... more
    Polymorphisms in microRNAs and their target sites can disrupt microRNA-dependent gene regulation, and have been associated with cancer susceptibility. However, genome-scale analyses of microRNA-related genetic variation in cancer are lacking. We tested the associations of ~40,000 common (MAF≥5%), microRNA-related single nucleotide polymorphisms (miR-SNPs), with risk of head and neck squamous cell carcinoma (HNSCC) in a discovery population, and validated selected loci in an independent population among a total of 2,198 cases and 2,180 controls. Joint analyses across the discovery and validation populations revealed six novel miR-SNP associations with risk of HNSCC. An upstream variant of MIR548H4 (rs7834169), replicated its association with overall HNSCC risk as well as risk of oral cavity cancer. Four other variants were specifically associated with oral cavity cancer risk (rs16914640, rs1134367, rs7306991 and rs1373756). 3'UTR variant of HADH, rs221347, and rs4975616, located ...
    E-Cadherin and N-cadherin are important components of epithelial-mesenchymal transition (EMT). The majority of studies on EMT in melanoma have been performed with cultured cell lines or pooled melanoma samples. The goal of our study was... more
    E-Cadherin and N-cadherin are important components of epithelial-mesenchymal transition (EMT). The majority of studies on EMT in melanoma have been performed with cultured cell lines or pooled melanoma samples. The goal of our study was to evaluate the expression of E-cadherin and N-cadherin in matched tissue samples from primary and metastatic sites of melanoma and to determine the correlation with survival outcome. We analyzed tissues from 42 melanoma primary lesions and their corresponding metastases, as well as 53 benign nevi, for expression levels of E-cadherin and N-cadherin using immunohistochemical methods. There were heterogenous expression patterns of E- and N-cadherin in both primary and metastatic melanomas. Overall, metastatic tumor showed a decrease in E-cadherin expression and an increase in N-cadherin expression compared to the primary tumor, although the difference did not reach statistical significance (p=0.24 and 0.28 respectively). A switch of membranous expressi...
    A portable near-infrared spectral tomography (NIRST) system was developed with simultaneous frequency domain (FD) and continuous-wave (CW) optical measurements for efficient characterization of breast cancer in a clinical oncology... more
    A portable near-infrared spectral tomography (NIRST) system was developed with simultaneous frequency domain (FD) and continuous-wave (CW) optical measurements for efficient characterization of breast cancer in a clinical oncology setting. Simultaneous FD and CW recordings were implemented to speed up acquisition to 3 minutes for all 9 wavelengths, spanning a range from 661nm to 1064nm. An adjustable interface was designed to fit various breast sizes and shapes. Spatial images of oxy- and deoxy-hemoglobin, water, lipid, and scattering components were reconstructed using a 2D FEM approach. The system was tested on a group of 10 normal subjects, who were examined bilaterally and the recovered optical images were compared to radiographic breast density. Significantly higher total hemoglobin and water were estimated in the high density relative to low density groups. One patient with invasive ductal carcinoma was also examined and the cancer region was characterized as having a contrast...
    The purpose of this study is to determine the diagnostically most important molecular biomarkers quantified by MR-guided Near-Infrared Spectral Tomography (NIRST) that distinguish malignant breast lesions from benign abnormalities when... more
    The purpose of this study is to determine the diagnostically most important molecular biomarkers quantified by MR-guided Near-Infrared Spectral Tomography (NIRST) that distinguish malignant breast lesions from benign abnormalities when combined with outcomes from clinical breast MRI. The study was HIPAA compliant and approved by the Dartmouth Institutional Review Board, the National Institutes of Health, the United States State Department, and Xijing Hospital. MR-guided NIRST evaluated hemoglobin, water, and lipid content in regions of interest defined by concurrent dynamic contrast-enhanced MRI (DCE-MRI) in the breast. MRI plus NIRST was performed in 44 subjects (median age: 46, age range: 20-81), 28 of whom had subsequent malignant pathological diagnoses, and 16 had benign conditions. A subset of 30 subject exams yielded optical data that met minimum sensitivity requirements to the suspicious lesion and were included in the analyses of diagnostic performance. In the subset of 30 s...
    ABSTRACT Analysis of NIR tomography imaging of 19 patients undergoing neoadjuvant chemotherapy shows pre-TX tumor HbT was statistically significant in differentiating the patients with and without complete pathological response... more
    ABSTRACT Analysis of NIR tomography imaging of 19 patients undergoing neoadjuvant chemotherapy shows pre-TX tumor HbT was statistically significant in differentiating the patients with and without complete pathological response (P-value=0.01, AUC=0.92).
    This study investigated the impact of prognostic variables, including the distance a patient lives from a transplant center, on the outcome of autologous hematopoietic stem cell transplantation (ASCT) for multiple myeloma. This... more
    This study investigated the impact of prognostic variables, including the distance a patient lives from a transplant center, on the outcome of autologous hematopoietic stem cell transplantation (ASCT) for multiple myeloma. This retrospective analysis included 77 myeloma patients who received an ASCT at Dartmouth Hitchcock Medical Center between 1996 and 2009, 70 of whom were treated between 2002 and 2009. Using linear regression and univariate analysis, we examined the impact of distance from the transplant center on survival. Kaplan-Meier curves identified overall and event-free survival. An association between distance from the transplant center and survival was examined using Cox regression analysis, while adjusting for patient-, disease-, and treatment-related variables. Increasing distance from the transplant center correlated with improved overall survival (P=.004), but had no impact on disease-free survival (P=.26). These results suggest that the distance from a transplant ce...
    To determine whether pretreatment biomarkers obtained from diffuse optical spectroscopic tomographic (DOST) imaging predicts breast tumor response to neoadjuvant chemotherapy (NAC), which would have value to potentially eliminate delays... more
    To determine whether pretreatment biomarkers obtained from diffuse optical spectroscopic tomographic (DOST) imaging predicts breast tumor response to neoadjuvant chemotherapy (NAC), which would have value to potentially eliminate delays in prescribing definitive local regional therapy that may occur from a standard complete 6- to 8-month course of NAC. Nineteen patients undergoing NAC were imaged with DOST before, during, and after treatment. The DOST images of total hemoglobin concentration (HbT), tissue oxygen saturation (StO2), and water (H2O) fraction at different time points have been used for testing the abilities of differentiating patients having pathologic complete response (pCR) versus pathologic incomplete response (pIR). Significant differences (P < 0.001, AUC = 1.0) were found between pCR patients versus pIR in outcome, based on the percentage change in tumor HbT within the first cycle of treatment. In addition, pretreatment tumor HbT (pretreatment HbT) relative to t...
    The effect of supplementation with calcium alone on risk fractures in a healthy population is not clear. The objective was to determine whether 4 y of calcium supplementation would reduce the fracture risk during treatment and subsequent... more
    The effect of supplementation with calcium alone on risk fractures in a healthy population is not clear. The objective was to determine whether 4 y of calcium supplementation would reduce the fracture risk during treatment and subsequent follow-up in a randomized placebo-controlled trial. The participants were aged <80 y at study entry (mean age: 61 y), were generally healthy, and had a recent diagnosis of colorectal adenoma. A total of 930 participants (72% men; mean age: 61 y) were randomly assigned to receive 4 y of treatment with 3 g CaCO(3) (1200 mg elemental Ca) daily or placebo and were followed for a mean of 10.8 y. The primary outcomes of this analysis were all fractures and minimal trauma fractures (caused by a fall from standing height or lower while sitting, standing, or walking). There were 46 fractures (15 from minimal trauma) in 464 participants in the calcium group and 54 (29 from minimal trauma) in 466 participants in the placebo group. The overall risk of fractu...
    To investigate an association between secondary polycythemia and venous thromboembolism (VTE) risk, we performed a case–control study to compare the prevalence of VTE in participants with secondary polycythemia due to chronic obstructive... more
    To investigate an association between secondary polycythemia and venous thromboembolism (VTE) risk, we performed a case–control study to compare the prevalence of VTE in participants with secondary polycythemia due to chronic obstructive pulmonary disease (COPD; N = 86) to that in age- and sex-matched controls with COPD without secondary polycythemia (N = 86). Although there was a significant difference in mean hematocrit between cases and controls (53.5% vs 43.6%, respectively; P < .005), we identified no difference in the number of total or idiopathic VTE events in the 2 groups. Patients with VTE, however, had a significantly higher body mass index than patients without VTE. Our findings suggest that secondary polycythemia alone may not be a significant risk factor for VTE but that VTE risk in this population may be related to known risk factors such as obesity. The role of phlebotomy for VTE risk reduction secondary polycythemia is therefore questionable.
    Given the density of single-nucleotide polymorphisms (SNP) in the human genome and the sensitivity of single-nucleotide changes in microRNA (miRNA) functionality and processing, we asked whether polymorphisms within miRNA processing... more
    Given the density of single-nucleotide polymorphisms (SNP) in the human genome and the sensitivity of single-nucleotide changes in microRNA (miRNA) functionality and processing, we asked whether polymorphisms within miRNA processing pathways and binding sites may influence non–small cell lung cancer (NSCLC) patients' prognosis. We genotyped 240 miRNA-related SNPs in 535 patients with stage I and II NSCLCs to determine associations with overall recurrence and survival as well as effect in specific treatment subgroups. After correcting for multiple comparisons, the G allele of FZD4:rs713065 displayed a significant association with decreased risk of death in surgery-only patients [HR, 0.46; 95% confidence interval (CI), 0.32–0.65]. DROSHA:rs6886834 variant A allele (HR, 6.38; 95% CI, 2.49–16.31) remained significant for increased risk of recurrence in the overall and surgery-only populations, respectively. FAS:rs2234978 G allele remained significantly associated with survival in al...
    Gene-specific promoter methylation of several genes occurs in aging normal tissues and may predispose to tumorigenesis. In the present study, we investigate the association of blood folate levels and dietary and lifestyle factors with CpG... more
    Gene-specific promoter methylation of several genes occurs in aging normal tissues and may predispose to tumorigenesis. In the present study, we investigate the association of blood folate levels and dietary and lifestyle factors with CpG island (CGI) methylation in normal colorectal mucosa. Subjects were enrolled in a multicenter chemoprevention trial of aspirin or folic acid for the prevention of large bowel adenomas. We collected 1,000 biopsy specimens from 389 patients, 501 samples from the right colon and 499 from the rectum at the follow-up colonoscopy. We measured DNA methylation of estrogen receptor alpha (ERα) and secreted frizzled related protein-1 (SFRP1), using bisulfite pyrosequencing. We used generalized estimating equations regression analysis to examine the association between methylation and selected variables. For both ERα and SFRP1, percentage methylation was significantly higher in the rectum than in the right colon (P = 0.001). For each 10 years of age, we obser...
    Some serrated polyps of the colorectum are likely preinvasive lesions, evolving through a newly recognized serrated pathway to colorectal cancer. To assess possible risk and protective factors for serrated polyps and particularly to... more
    Some serrated polyps of the colorectum are likely preinvasive lesions, evolving through a newly recognized serrated pathway to colorectal cancer. To assess possible risk and protective factors for serrated polyps and particularly to explore differences in risk factors between polyps in the right and left colorectum, we pooled data from three large multicenter chemoprevention trials. A serrated polyp was defined broadly as any serrated lesion (hyperplastic, sessile serrated adenoma, “traditional” serrated adenoma, mixed adenoma) diagnosed during each trial's main treatment period of ∼3 to 4 years. Using generalized linear regression, we computed risk ratios and 95% confidence intervals as measures of the association between risk for serrated polyps and demographic, lifestyle, and dietary variables. Of the 2,830 subjects that completed at least one follow-up exam after randomization, 675 (23.9%) had at least one left-sided serrated polyp and 261 (9.2%) had at least one right-sided...
    Genome-wide association studies (GWAS) have proven successful in predicting genetic risk of disease using single-locus models; however, identifying single nucleotide polymorphism (SNP) interactions at the genome-wide scale is limited due... more
    Genome-wide association studies (GWAS) have proven successful in predicting genetic risk of disease using single-locus models; however, identifying single nucleotide polymorphism (SNP) interactions at the genome-wide scale is limited due to computational and statistical challenges. We address the computational burden encountered when detecting SNP interactions for survival analysis, such as age of disease-onset. To confront this problem, we developed a novel algorithm, called the Efficient Survival Multifactor Dimensionality Reduction (ES-MDR) method, which uses Martingale Residuals as the outcome parameter to estimate survival outcomes, and implemented the Quantitative Multifactor Dimensionality Reduction method to identify significant interactions associated with age of disease-onset. To demonstrate efficacy, we evaluated this method on two simulation sets to estimate the type I error and power. Simulations show that ES-MDR identifies interactions using less computational workload...
    An important area of research in pharmacogenomics is to relate high-dimensional genetic or genomic data to various clinical phenotypes of patients. Due to large variability in time to certain clinical event among patients, studying... more
    An important area of research in pharmacogenomics is to relate high-dimensional genetic or genomic data to various clinical phenotypes of patients. Due to large variability in time to certain clinical event among patients, studying possibly cen- sored survival phenotypes can be more informative than treating the phenotypes as categorical variables. In this paper, we develop a threshold gradient descent (TGD)
    The purpose of this study is to determine the diagnostically most important molecular biomarkers quantified by MR-guided Near-Infrared Spectral Tomography (NIRST) that distinguish malignant breast lesions from benign abnormalities when... more
    The purpose of this study is to determine the diagnostically most important molecular biomarkers quantified by MR-guided Near-Infrared Spectral Tomography (NIRST) that distinguish malignant breast lesions from benign abnormalities when combined with outcomes from clinical breast MRI. The study was HIPAA compliant and approved by the Dartmouth Institutional Review Board, the National Institutes of Health, the United States State Department, and Xijing Hospital. MR-guided NIRST evaluated hemoglobin, water, and lipid content in regions of interest defined by concurrent dynamic contrast-enhanced MRI (DCE-MRI) in the breast. MRI plus NIRST was performed in 44 subjects (median age: 46, age range: 20-81), 28 of whom had subsequent malignant pathological diagnoses, and 16 had benign conditions. A subset of 30 subject exams yielded optical data that met minimum sensitivity requirements to the suspicious lesion and were included in the analyses of diagnostic performance. In the subset of 30 s...
    An optimized approach to nonlinear iterative reconstruction of magnetic resonance imaging (MRI)-guided near-infrared spectral tomography (NIRST) images was developed using an L-curve-based algorithm for the choice of regularization... more
    An optimized approach to nonlinear iterative reconstruction of magnetic resonance imaging (MRI)-guided near-infrared spectral tomography (NIRST) images was developed using an L-curve-based algorithm for the choice of regularization parameter. This approach was applied to clinical exam data to maximize the reconstructed values differentiating malignant and benign lesions. MRI/NIRST data from 25 patients with abnormal breast readings (BI-RADS category 4-5) were analyzed using this optimal regularization methodology, and the results showed enhanced p values and area under the curve (AUC) for the task of differentiating malignant from benign lesions. Of the four absorption parameters and two scatter parameters, the most significant differences for benign versus malignant were total hemoglobin (HbT) and tissue optical index (TOI) with p values = 0.01 and 0.001, and AUC values = 0.79 and 0.94, respectively, in terms of HbT and TOI. This dramatically improved the values relative to fixed regularization (p value = 0.02 and 0.003; AUC = 0.75 and 0.83) showing that more differentiation was possible with the optimal method. Through a combination of both biomarkers, HbT and TOI, the AUC increased from 82.9% (fixed regulation = 0.1) to 94.3% (optimal method).
    ABSTRACT Analysis of NIR tomography imaging of 19 patients undergoing neoadjuvant chemotherapy shows pre-TX tumor HbT was statistically significant in differentiating the patients with and without complete pathological response... more
    ABSTRACT Analysis of NIR tomography imaging of 19 patients undergoing neoadjuvant chemotherapy shows pre-TX tumor HbT was statistically significant in differentiating the patients with and without complete pathological response (P-value=0.01, AUC=0.92).
    To determine whether pretreatment biomarkers obtained from diffuse optical spectroscopic tomographic (DOST) imaging predicts breast tumor response to neoadjuvant chemotherapy (NAC), which would have value to potentially eliminate delays... more
    To determine whether pretreatment biomarkers obtained from diffuse optical spectroscopic tomographic (DOST) imaging predicts breast tumor response to neoadjuvant chemotherapy (NAC), which would have value to potentially eliminate delays in prescribing definitive local regional therapy that may occur from a standard complete 6- to 8-month course of NAC. Nineteen patients undergoing NAC were imaged with DOST before, during, and after treatment. The DOST images of total hemoglobin concentration (HbT), tissue oxygen saturation (StO2), and water (H2O) fraction at different time points have been used for testing the abilities of differentiating patients having pathologic complete response (pCR) versus pathologic incomplete response (pIR). Significant differences (P < 0.001, AUC = 1.0) were found between pCR patients versus pIR in outcome, based on the percentage change in tumor HbT within the first cycle of treatment. In addition, pretreatment tumor HbT (pretreatment HbT) relative to t...
    The 15q25.1 lung cancer susceptibility locus, containing CHRNA5, could modify lung cancer susceptibility and multiple smoking related phenotypes. However, no studies have investigated the association between CHRNA5 rs3841324, which has... more
    The 15q25.1 lung cancer susceptibility locus, containing CHRNA5, could modify lung cancer susceptibility and multiple smoking related phenotypes. However, no studies have investigated the association between CHRNA5 rs3841324, which has been proven to have the highest association with CHRNA5 mRNA expression, and the risk of other smoking-associated cancers, except lung cancer. In the current study we examined the association between rs3841324 and susceptibility to smoking-associated nasopharyngeal carcinoma (NPC). In this case-control study we genotyped the CHRNA5 rs3841324 polymorphism with 400 NPC cases and 491 healthy controls who were Han Chinese and frequency-matched by age (±5 years), gender, and alcohol consumption. Univariate and multivariate logistic regression analyses were used to calculate the odds ratio (OR) and 95% confidence intervals (95% CI). We found that individuals with CHRNA5 rs3841324 combined variant genotypes (ins/del+del/del) had a >1.5-fold elevated risk ...
    ABSTRACT In genome-wide studies, hundreds of thousands of hypothesis tests are performed simultaneously. Bonferroni correction and False Discovery Rate (FDR) can effectively control type I error but often yield a high false negative rate.... more
    ABSTRACT In genome-wide studies, hundreds of thousands of hypothesis tests are performed simultaneously. Bonferroni correction and False Discovery Rate (FDR) can effectively control type I error but often yield a high false negative rate. We aim to develop a more powerful method to detect differentially expressed genes. We present a Weighted False Discovery Rate (WFDR) method that incorporate biological knowledge from genetic networks. We first identify weights using Integrative Multi-species Prediction (IMP) and then apply the weights in WFDR to identify differentially expressed genes through an IMP-WFDR algorithm. We performed a gene expression experiment to identify zebrafish genes that change expression in the presence of arsenic during a systemic Pseudomonas aeruginosa infection. Zebrafish were exposed to arsenic at 10 parts per billion and/or infected with P. aeruginosa. Appropriate controls were included. We then applied IMP-WFDR during the analysis of differentially expressed genes. We compared the mRNA expression for each group and found over 200 differentially expressed genes and several enriched pathways including defense response pathways, arsenic response pathways, and the Notch signaling pathway.
    Regulatory T cells (Tregs) are potentially prognostic indicators in patients with glioblastoma. If differences in frequency of Tregs in tumor or blood account for substantial variation in patient survival, then reliably measuring Tregs... more
    Regulatory T cells (Tregs) are potentially prognostic indicators in patients with glioblastoma. If differences in frequency of Tregs in tumor or blood account for substantial variation in patient survival, then reliably measuring Tregs may enhance treatment selection and improve outcomes. We measured Tregs and CD3+ T cells in tumors and blood from 25 patients with newly diagnosed glioblastoma. Tumor-infiltrating Tregs and CD3+ T cells, measured by quantitative DNA demethylation analysis (epigenetic qPCR) and by immunohistochemistry, and peripheral blood Treg proportions measured by flow cytometry were correlated with patient survival. Additionally, we analyzed data from The Cancer Genome Atlas (TCGA) to correlate the expression of Treg markers with patient survival and glioblastoma subtypes. Tregs, as measured in tumor tissue and peripheral blood, did not correlate with patient survival. Although there was a correlation between tumor-infiltrating Tregs expression by epigenetic qPCR ...
    SUMMARY Large-scale microarray gene expression data provide the possibility of constructing genetic networks or biological pathways. Gaussian graphical models have been suggested to provide an effective method for constructing such... more
    SUMMARY Large-scale microarray gene expression data provide the possibility of constructing genetic networks or biological pathways. Gaussian graphical models have been suggested to provide an effective method for constructing such genetic networks. However, most of the available methods for constructing Gaussian graphs do not account for the sparsity of the networks and are computationally more demanding or in- feasible, especially in the settings of high dimension and low sample size. We introduce a threshold gradient descent (TGD) regularization procedure for estimating the sparse precision matrix in the setting of Gaussian graphical models and demonstrate its application to identifying genetic networks. Such a procedure is computationally feasible and can easily incorporate prior biological knowledge about the network structure. Simulation results indicate that the proposed method yields a better estimate of the precision matrix than the procedures that fail to account for the s...
    There is an imperative need to develop methods that can rapidly and accurately determine individual exposure to radiation for screening (triage) populations and guiding medical treatment in an emergency response to a large-scale... more
    There is an imperative need to develop methods that can rapidly and accurately determine individual exposure to radiation for screening (triage) populations and guiding medical treatment in an emergency response to a large-scale radiological/nuclear event. To this end, a number of methods that rely on dose-dependent chemical and/or physical alterations in biomaterials or biological responses are in various stages of development. One such method, ex vivo electron paramagnetic resonance (EPR) nail dosimetry using human nail clippings, is a physical biodosimetry technique that takes advantage of a stable radiation-induced signal (RIS) in the keratin matrix of fingernails and toenails. This dosimetry method has the advantages of ubiquitous availability of the dosimetric material, easy and non-invasive sampling, and the potential for immediate and rapid dose assessment. The major challenge for ex vivo EPR nail dosimetry is the overlap of mechanically induced signals and the RIS. The difficulties of analysing the mixed EPR spectra of a clipped irradiated nail were addressed in the work described here. The following key factors lead to successful spectral analysis and dose assessment in ex vivo EPR nail dosimetry: (1) obtaining a thorough understanding of the chemical nature, the decay behaviour, and the microwave power dependence of the EPR signals, as well as the influence of variation in temperature, humidity, water content, and O₂ level; (2) control of the variability among individual samples to achieve consistent shape and kinetics of the EPR spectra; (3) use of correlations between the multiple spectral components; and (4) use of optimised modelling and fitting of the EPR spectra to improve the accuracy and precision of the dose estimates derived from the nail spectra. In the work described here, two large clipped nail datasets were used to test the procedures and the spectral fitting model of the results obtained with it. A 15-donor nail set with 90 nail samples from 15 donors was used to validate the sample handling and spectral analysis methods that have been developed but without the interference of a native background signal. Good consistency has been obtained between the actual RIS and the estimated RIS computed from spectral analysis. In addition to the success in RIS estimation, a linear dose response has also been achieved for all individuals in this study, where the radiation dose ranges from 0 to 6 Gy. A second 16-donor nail set with 96 nail samples was used to test the spectral fitting model where the background signal was included during the fitting of the clipped nail spectra data. Although the dose response for the estimated and actual RIS calculated in both donor nail sets was similar, there was an increased variability in the RIS values that was likely due to the variability in the background signal between donors. Although the current methods of sample handling and spectral analysis show good potential for estimating the RIS in the EPR spectra of nail clippings, there is a remaining degree of variability in the RIS estimate that needs to be addressed; this should be achieved by identifying and accounting for demographic sources of variability in the background nail signal and the composition of the nail matrix.
    Distinguishing cellular abnormalities in reactive and malignant lesions is challenging. We compared the incidence and severity of cytological abnormalities in malignant/premalignant and benign epidermal lesions. One hundred fifty-two... more
    Distinguishing cellular abnormalities in reactive and malignant lesions is challenging. We compared the incidence and severity of cytological abnormalities in malignant/premalignant and benign epidermal lesions. One hundred fifty-two biopsies representing 69 malignant/premalignant squamous lesions and 83 benign conditions were studied. Cytological features, including nuclear hyperchromasia, nuclear overlap (crowding), irregular nuclei, high nuclear/cytoplasmic (N/C) ratio, conspicuous nucleoli, delicate inconspicuous nucleoli, clumped chromatin, pleomorphic parakeratosis, normal and abnormal mitotic figures and necrotic keratinocytes, were evaluated and graded. Statistical analysis was performed. Irregular nuclei, increased N/C ratio, conspicuous single prominent nucleoli, nuclear overlap (crowding), pleomorphic parakeratosis, nuclear hyperchromasia, necrotic keratinocytes, normal and abnormal mitotic figures and coarse chromatin were seen more frequently in malignant neoplasms (p < 0.05). Abnormal mitotic figures, although uncommon (20.3%), were only noted in the malignant/premalignant group. Certain cytological features were common among both malignant and benign lesions, suggesting that they are of little value. In the setting of an atypical cutaneous squamous proliferation, nuclear irregularity, increased N/C ratio, conspicuous nucleoli, crowding and hyperchromasia are the most useful indicators of malignancy. In contrast, mitotic figures, necrotic cells and coarse chromatin are less useful. The presence of abnormal mitotic figures is very helpful when present; however, their overall rarity limits their utility.
    Basal cell carcinoma (BCC) has a characteristic stroma, but less is known about the dermal characteristics associated with melanoma in situ (MIS) and actinic keratosis (AK). Dermal changes were studied in 301 specimens of AK, BCC and MIS.... more
    Basal cell carcinoma (BCC) has a characteristic stroma, but less is known about the dermal characteristics associated with melanoma in situ (MIS) and actinic keratosis (AK). Dermal changes were studied in 301 specimens of AK, BCC and MIS. Subsequently, blinded images of dermal changes from 90 randomly selected cases of those entities were used to assess the predictive value of the dermal changes. Agreement with the final diagnosis was calculated using kappa coefficient (κ). Fibromyxoid stroma was present in 82% of BCC cases; fibrous stroma was seen in 25% of BCC, 58% of MIS and 35.6% of AK specimens (p < 0.05). A lichenoid inflammatory infiltrate was frequently associated with AK and a perifollicular infiltrate with periadnexal fibrosis with MIS. Blinded evaluation of images of the dermal changes associated with the tumors yielded the correct diagnosis in (54.4, 41.1 and 27.8%; average 41.2%) by the three appraisers. Coefficient of agreement in blinded imaged evaluation with the actual diagnosis was higher in the BCC and MIS compared with AK (κ = 0.37, p = 0.0001; κ = 0.2, p = 0.0005 and κ = -0.06, p = 0.84, respectively). Dermal features may be helpful in predicting the correct diagnosis when tumor is not visible.
    The workhorse of modern genetic analysis is the parametric linear model. The advantages of the linear modeling framework are many and include a mathematical understanding of the model fitting process and ease of interpretation. However,... more
    The workhorse of modern genetic analysis is the parametric linear model. The advantages of the linear modeling framework are many and include a mathematical understanding of the model fitting process and ease of interpretation. However, an important limitation is that linear models make assumptions about the nature of the data being modeled. This assumption may not be realistic for complex biological systems such as disease susceptibility where nonlinearities in the genotype to phenotype mapping relationship that result from epistasis, plastic reaction norms, locus heterogeneity, and phenocopy, for example, are the norm rather than the exception. We have previously developed a flexible modeling approach called symbolic discriminant analysis (SDA) that makes no assumptions about the patterns in the data. Rather, SDA lets the data dictate the size, shape, and complexity of a symbolic discriminant function that could include any set of mathematical functions from a list of candidates supplied by the user. Here, we outline a new five step process for symbolic model discovery that uses genetic programming (GP) for coarse-grained stochastic searching, experimental design for parameter optimization, graphical modeling for generating expert knowledge, and estimation of distribution algorithms for fine-grained stochastic searching. Finally, we introduce function mapping as a new method for interpreting symbolic discriminant functions. We show that function mapping when combined with measures of interaction information facilitates statistical interpretation by providing a graphical approach to decomposing complex models to highlight synergistic, redundant, and independent effects of polymorphisms and their composite functions. We illustrate this five step SDA modeling process with a real case-control dataset.

    And 7 more