New algorithms for automated phylogenetic reconstruction using artificial intelligence and data mining techniques

Academic and industrial research in the late 90s have brought about an exponential explosion of DNA sequence data. Automated expert systems are being created to help biologists to extract patterns, trends and links from this ever-deepening ocean of information. Two such systems aimed on retrieving and subsequently utilizing phylogenetically relevant information have been developed in this dissertation, the major objective of which was to automate the often difficult and confusing phylogenetic reconstruction process. ^ Popular phylogenetic reconstruction methods, such as distance-based methods, attempt to find an optimal tree topology (that reflects the relationships among related sequences and their evolutionary history) by searching through the topology space. Various compromises between the fast (but incomplete) and exhaustive (but computationally prohibitive) search heuristics have been suggested. An intelligent compromise algorithm that relies on a flexible “beam” search principle from the Artificial Intelligence domain and uses the pre-computed local topology reliability information to adjust the beam search space continuously is described in the second chapter of this dissertation. ^ However, sometimes even a (virtually) complete distance-based method is inferior to the significantly more elaborate (and computationally expensive) maximum likelihood (ML) method. In fact, depending on the nature of the sequence data in question either method might prove to be superior. Therefore, it is difficult (even for an expert) to tell a priori which phylogenetic reconstruction method—distance-based, ML or maybe maximum parsimony (MP)—should be chosen for any particular data set. ^ A number of factors, often hidden, influence the performance of a method. For example, it is generally understood that for a phylogenetically “difficult” data set more sophisticated methods (e.g., ML) tend to be more effective and thus should be chosen. However, it is the interplay of many factors that one needs to consider in order to avoid choosing an inferior method (potentially a costly mistake, both in terms of computational expenses and in terms of reconstruction accuracy.) ^ Chapter III of this dissertation details a phylogenetic reconstruction expert system that selects a superior proper method automatically. It uses a classifier (a Decision Tree-inducing algorithm) to map a new data set to the proper phylogenetic reconstruction method. ^

Annexin A5-Conjugated Polymeric Micelles For Dual SPECT and Optical Detection of Apoptosis

Apoptosis, a form of programmed cell death, is critical to homoeostasis, normal development, and physiology. Dysregulation of apoptosis can lead to the accumulation of unwanted cells, such as occurs in cancer, and the removal of needed cells or disorders of normal tissues, such as heart, neurodegenerative, and autoimmune diseases. Noninvasive detection of apoptosis may play an important role in the evaluation of disease states and response to therapeutic intervention for a variety of diseases. It is desirable to have an imaging method to accurately detect and monitor this process in patients. In this study, we developed annexin A5-conjugated polymeric micellar nanoparticles dual-labeled with a near-infrared fluorescence fluorophores (Cy7) and a radioisotope (111In), named as 111In-labeled annexin A5-CCPM. In vitro studies demonstrated that annexin A5-CCPM could strongly and specifically bind to apoptotic cells. In vivo studies showed that apoptotic tissues could be clearly visualized by both single photon emission computed tomography (SPECT) and fluorescence molecular tomography (FMT) after intravenous injection of 111In-labeled Annexin A5-CCPM in 6 different apoptosis models. In contrast, there was little signal in respective healthy tissues. All the biodistribution data confirmed imaging results. Moreover, histological analysis revealed that radioactivity count correlated with fluorescence signal from the nanoparticles, and both signals co-localized with the region of apoptosis. In sum, 111In-labeled annexin A5-CCPM allowed visualization of apoptosis by both nuclear and optical imaging techniques. The complementary information acquired with multiple imaging techniques should be advantageous in improving diagnostics and management of patients.

Cytochrome P450 in mammalian brain: Purification, anatomical distribution and regulation of the cytochrome P450 monooxygenase system in rat brain

The cytochrome P450 (P450) monooxygenase system plays a major role in metabolizing a wide variety of xenobiotic as well as endogenous compounds. In performing this function, it serves to protect the body from foreign substances. However, in a number of cases, P450 activates procarcinogens to cause harm. In most animals, the highest level of activity is found in the liver. Virtually all tissues demonstrate P450 activity, though, and the role of the P450 monooxygenase system in these other organs is not well understood. In this project I have studied the P450 system in rat brain; purifying NADPH-cytochrome P450 reductase (reductase) from that tissue. In addition, I have examined the distribution and regulation of expression of reductase and P450 in various anatomical regions of the rat brain.^ NADPH-cytochrome P450 reductase was purified to apparent homogeneity and cytochrome P450 partially purified from whole rat brain. Purified reductase from brain was identical to liver P450 reductase by SDS-PAGE and Western blot techniques. Kinetic studies utilizing cerebral P450 reductase reveal Km values in close agreement with those determined with enzyme purified from rat liver. Moreover, the brain P450 reductase was able to function successfully in a reconstituted microsomal system with partially purified brain cytochrome P450 and with purified hepatic P4501A1 as measured by 7-ethoxycoumarin and 7-ethoxyresorufin O-deethylation. These results indicate that the reductase and P450 components may interact to form a competent drug metabolism system in brain tissue.^ Since the brain is not a homogeneous organ, dependent upon the well orchestrated interaction of numerous parts, pathology in one nucleus may have a large impact upon its overall function. Hence, the anatomical distribution of the P450 monooxygenase system in brain is important in elucidating its function in that organ. Related to this is the regulation of P450 expression in brain. In order to study these issues female rats--both ovariectomized and not--were treated with a number of xenobiotic compounds and sex steroids. The brains from these animals were dissected into 8 discrete regions and the presence and relative level of message for P4502D and reductase determined using polymerase chain reaction. Results of this study indicate the presence of mRNA for reductase and P4502D isoforms throughout the rat brain. In addition, quantitative PCR has allowed the determination of factors affecting the expression of message for these enzymes. ^

Development and formative evaluation of an obesity prevention intervention for the pediatric primary care setting: Helping HAND (Healthy Activity and Nutrition Directions)

As the obesity epidemic continues to increase, the pediatric primary care office setting remains a relatively unexplored arena to offer obesity prevention interventions for children. The increased risk for adult obesity among 10 to 14 year-old children who are overweight, suggests obesity prevention programs should be introduced just before this age or early in this age period. Research is also accumulating on the importance of targeting parents along with children, since parents are in charge of the home environment for children. Therefore, the aim of this project was to develop an obesity prevention program called Helping HAND (Healthy Activity and Nutrition Directions) based on Social Cognitive Theory and authoritative parenting techniques for the pediatric primary care setting and conduct one-on-one interviews with parents as the initial formative evaluation of the intervention material for the obesity prevention intervention. A secondary aim of the project was to determine the feasibility of identifying appropriate subjects for the intervention, and conducting qualitative evaluations of the materials through recruitment through pediatric primary care settings. ^

Toxics Release Inventory facilities and childhood cancer: Geographic information systems based approach

Purpose. To examine the association between living in proximity to Toxics Release Inventory (TRI) facilities and the incidence of childhood cancer in the State of Texas. ^ Design. This is a secondary data analysis utilizing the publicly available Toxics release inventory (TRI), maintained by the U.S. Environmental protection agency that lists the facilities that release any of the 650 TRI chemicals. Total childhood cancer cases and childhood cancer rate (age 0-14 years) by county, for the years 1995-2003 were used from the Texas cancer registry, available at the Texas department of State Health Services website. Setting: This study was limited to the children population of the State of Texas. ^ Method. Analysis was done using Stata version 9 and SPSS version 15.0. Satscan was used for geographical spatial clustering of childhood cancer cases based on county centroids using the Poisson clustering algorithm which adjusts for population density. Pictorial maps were created using MapInfo professional version 8.0. ^ Results. One hundred and twenty five counties had no TRI facilities in their region, while 129 facilities had at least one TRI facility. An increasing trend for number of facilities and total disposal was observed except for the highest category based on cancer rate quartiles. Linear regression analysis using log transformation for number of facilities and total disposal in predicting cancer rates was computed, however both these variables were not found to be significant predictors. Seven significant geographical spatial clusters of counties for high childhood cancer rates (p<0.05) were indicated. Binomial logistic regression by categorizing the cancer rate in to two groups (<=150 and >150) indicated an odds ratio of 1.58 (CI 1.127, 2.222) for the natural log of number of facilities. ^ Conclusion. We have used a unique methodology by combining GIS and spatial clustering techniques with existing statistical approaches in examining the association between living in proximity to TRI facilities and the incidence of childhood cancer in the State of Texas. Although a concrete association was not indicated, further studies are required examining specific TRI chemicals. Use of this information can enable the researchers and public to identify potential concerns, gain a better understanding of potential risks, and work with industry and government to reduce toxic chemical use, disposal or other releases and the risks associated with them. TRI data, in conjunction with other information, can be used as a starting point in evaluating exposures and risks. ^

Quality of life, symptom experience, and use of complementary and alternative medicine among women with ovarian cancer

The objectives of this dissertation were to determine the quality of life in women with ovarian cancer and the association of their physical and emotional well-being with the number of symptoms, duration of symptoms, and the scores of common symptoms of ovarian cancer; to study the prevalence of complementary and alternative medicine techniques for symptom relief and its association with the number of symptoms, age, education, insurance, comorbidity, and satisfaction with medical care they received, and their pre-diagnostic experience of symptoms.^ This study was based on a secondary data analysis of a study of early detection of ovarian cancer. A sample of 139 women with ovarian cancer was recruited and was administered a questionnaire comprised of questions on their quality of life, their symptoms and what they did about the symptoms, whether they used any complementary and alternative medicine techniques, and other medical conditions they had. Out of this sample, 53 patients underwent in-depth interviews relating to their symptoms before the diagnosis and their experiences with the health care system leading to the ovarian cancer diagnosis. ^ In article #1, ovarian cancer patients were observed to have significantly poorer quality of life on all subscales and summary scores except pain, compared to that of the general population of US women. Physical well-being scores were negatively associated with the number of symptoms before diagnosis and a significant negative association of comorbidity index was observed with physical well-being. Higher education and increase in time since diagnosis was found to have better physical scores. Emotional well-being scores showed marginally significant associations with number of symptoms and bloating. ^ In article #2, a thematic content analysis of the ovarian cancer patients’ interviews revealed that on recognition of their symptoms women first assumed their symptoms to be a normal transient occurrence due to a pre-existing disease condition, or due to some other disease. A series of misattributions of their symptoms on their and their doctors’ part impacted their health care seeking.In article #3, a significantly greater likelihood of CAM use with an increase in the number of symptoms was observed.^ Based on the foregoing results, it is important to educate women on possible signs of ovarian cancer and also to educate doctors about the results of current research regarding ovarian cancer diagnosis. This will help to avoid a delay in getting a diagnosis and improve women’s quality of life. It emphasizes the diagnosis of ovarian cancer in earlier stages by more sensitive screening techniques. This study emphasizes the importance of consideration of comorbidity in any quality of life research. Additionally, educating women in the safe use of CAM techniques carries immense significance because the efficacy and safety of many of the currently advertized CAM products has not been scientifically validated. Further research is needed to confirm the findings of this study. ^

A comparison of techniques for forecasting utilization of Veterans Affairs hospitals

This study demonstrated that accurate, short-term forecasts of Veterans Affairs (VA) hospital utilization can be made using the Patient Treatment File (PTF), the inpatient discharge database of the VA. Accurate, short-term forecasts of two years or less can reduce required inventory levels, improve allocation of resources, and are essential for better financial management. These are all necessary achievements in an era of cost-containment.^ Six years of non-psychiatric discharge records were extracted from the PTF and used to calculate four indicators of VA hospital utilization: average length of stay, discharge rate, multi-stay rate (a measure of readmissions) and days of care provided. National and regional levels of these indicators were described and compared for fiscal year 1984 (FY84) to FY89 inclusive.^ Using the observed levels of utilization for the 48 months between FY84 and FY87, five techniques were used to forecast monthly levels of utilization for FY88 and FY89. Forecasts were compared to the observed levels of utilization for these years. Monthly forecasts were also produced for FY90 and FY91.^ Forecasts for days of care provided were not produced. Current inpatients with very long lengths of stay contribute a substantial amount of this indicator and it cannot be accurately calculated.^ During the six year period between FY84 and FY89, average length of stay declined substantially, nationally and regionally. The discharge rate was relatively stable, while the multi-stay rate increased slightly during this period. FY90 and FY91 forecasts show a continued decline in the average length of stay, while the discharge rate is forecast to decline slightly and the multi-stay rate is forecast to increase very slightly.^ Over a 24 month ahead period, all three indicators were forecast within a 10 percent average monthly error. The 12-month ahead forecast errors were slightly lower. Average length of stay was less easily forecast, while the multi-stay rate was the easiest indicator to forecast.^ No single technique performed significantly better as determined by the Mean Absolute Percent Error, a standard measure of error. However, Autoregressive Integrated Moving Average (ARIMA) models performed well overall and are recommended for short-term forecasting of VA hospital utilization. ^

Novel Imaging-Based Techniques Reveal a Role for PD-1/PD-L1 in Tumor Immune Surveillance in the Lung

The binding of immune inhibitory receptor Programmed Death 1 (PD-1) on T cells to its ligand PD-L1 has been implicated as a major contributor to tumor induced immune suppression. Clinical trials of PD-L1 blockade have proven effective in unleashing therapeutic anti-tumor immune responses in a subset of patients with advanced melanoma, yet current response rates are low for reasons that remain unclear. Hypothesizing that the PD-1/PD-L1 pathway regulates T cell surveillance within the tumor microenvironment, we employed intravital microscopy to investigate the in vivo impact of PD-L1 blocking antibody upon tumor-associated immune cell migration. However, current analytical methods of intravital dynamic microscopy data lack the ability to identify cellular targets of T cell interactions in vivo, a crucial means for discovering which interactions are modulated by therapeutic intervention. By developing novel imaging techniques that allowed us to better analyze tumor progression and T cell dynamics in the microenvironment; we were able to explore the impact of PD-L1 blockade upon the migratory properties of tumor-associated immune cells, including T cells and antigen presenting cells, in lung tumor progression. Our results demonstrate that early changes in tumor morphology may be indicative of responsiveness to anti-PD-L1 therapy. We show that immune cells in the tumor microenvironment as well as tumors themselves express PD-L1, but immune phenotype alone is not a predictive marker of effective anti-tumor responses. Through a novel method in which we quantify T cell interactions, we show that T cells are largely engaged in interactions with dendritic cells in the tumor microenvironment. Additionally, we show that during PD-L1 blockade, non-activated T cells are recruited in greater numbers into the tumor microenvironment and engage more preferentially with dendritic cells. We further show that during PD-L1 blockade, activated T cells engage in more confined, immune synapse-like interactions with dendritic cells, as opposed to more dynamic, kinapse-like interactions with dendritic cells when PD-L1 is free to bind its receptor. By advancing the contextual analysis of anti-tumor immune surveillance in vivo, this study implicates the interaction between T cells and tumor-associated dendritic cells as a possible modulator in targeting PD-L1 for anti-tumor immunotherapy.

Comparison of machine learning approaches and statistical methods for estimation of airborne chemical exposures

Accurate quantitative estimation of exposure using retrospective data has been one of the most challenging tasks in the exposure assessment field. To improve these estimates, some models have been developed using published exposure databases with their corresponding exposure determinants. These models are designed to be applied to reported exposure determinants obtained from study subjects or exposure levels assigned by an industrial hygienist, so quantitative exposure estimates can be obtained. ^ In an effort to improve the prediction accuracy and generalizability of these models, and taking into account that the limitations encountered in previous studies might be due to limitations in the applicability of traditional statistical methods and concepts, the use of computer science- derived data analysis methods, predominantly machine learning approaches, were proposed and explored in this study. ^ The goal of this study was to develop a set of models using decision trees/ensemble and neural networks methods to predict occupational outcomes based on literature-derived databases, and compare, using cross-validation and data splitting techniques, the resulting prediction capacity to that of traditional regression models. Two cases were addressed: the categorical case, where the exposure level was measured as an exposure rating following the American Industrial Hygiene Association guidelines and the continuous case, where the result of the exposure is expressed as a concentration value. Previously developed literature-based exposure databases for 1,1,1 trichloroethane, methylene dichloride and, trichloroethylene were used. ^ When compared to regression estimations, results showed better accuracy of decision trees/ensemble techniques for the categorical case while neural networks were better for estimation of continuous exposure values. Overrepresentation of classes and overfitting were the main causes for poor neural network performance and accuracy. Estimations based on literature-based databases using machine learning techniques might provide an advantage when they are applied to other methodologies that combine `expert inputs' with current exposure measurements, like the Bayesian Decision Analysis tool. The use of machine learning techniques to more accurately estimate exposures from literature-based exposure databases might represent the starting point for the independence from the expert judgment.^