TIME SERIES ANALYSIS AS INPUT FOR PREDICTIVE MODELING: PREDICTING CARDIAC ARREST IN A PEDIATRIC INTENSIVE CARE UNIT

The first manuscript, entitled "Time-Series Analysis as Input for Clinical Predictive Modeling: Modeling Cardiac Arrest in a Pediatric ICU" lays out the theoretical background for the project. There are several core concepts presented in this paper. First, traditional multivariate models (where each variable is represented by only one value) provide single point-in-time snapshots of patient status: they are incapable of characterizing deterioration. Since deterioration is consistently identified as a precursor to cardiac arrests, we maintain that the traditional multivariate paradigm is insufficient for predicting arrests. We identify time series analysis as a method capable of characterizing deterioration in an objective, mathematical fashion, and describe how to build a general foundation for predictive modeling using time series analysis results as latent variables. Building a solid foundation for any given modeling task involves addressing a number of issues during the design phase. These include selecting the proper candidate features on which to base the model, and selecting the most appropriate tool to measure them. We also identified several unique design issues that are introduced when time series data elements are added to the set of candidate features. One such issue is in defining the duration and resolution of time series elements required to sufficiently characterize the time series phenomena being considered as candidate features for the predictive model. Once the duration and resolution are established, there must also be explicit mathematical or statistical operations that produce the time series analysis result to be used as a latent candidate feature. In synthesizing the comprehensive framework for building a predictive model based on time series data elements, we identified at least four classes of data that can be used in the model design. The first two classes are shared with traditional multivariate models: multivariate data and clinical latent features. Multivariate data is represented by the standard one value per variable paradigm and is widely employed in a host of clinical models and tools. These are often represented by a number present in a given cell of a table. Clinical latent features derived, rather than directly measured, data elements that more accurately represent a particular clinical phenomenon than any of the directly measured data elements in isolation. The second two classes are unique to the time series data elements. The first of these is the raw data elements. These are represented by multiple values per variable, and constitute the measured observations that are typically available to end users when they review time series data. These are often represented as dots on a graph. The final class of data results from performing time series analysis. This class of data represents the fundamental concept on which our hypothesis is based. The specific statistical or mathematical operations are up to the modeler to determine, but we generally recommend that a variety of analyses be performed in order to maximize the likelihood that a representation of the time series data elements is produced that is able to distinguish between two or more classes of outcomes. The second manuscript, entitled "Building Clinical Prediction Models Using Time Series Data: Modeling Cardiac Arrest in a Pediatric ICU" provides a detailed description, start to finish, of the methods required to prepare the data, build, and validate a predictive model that uses the time series data elements determined in the first paper. One of the fundamental tenets of the second paper is that manual implementations of time series based models are unfeasible due to the relatively large number of data elements and the complexity of preprocessing that must occur before data can be presented to the model. Each of the seventeen steps is analyzed from the perspective of how it may be automated, when necessary. We identify the general objectives and available strategies of each of the steps, and we present our rationale for choosing a specific strategy for each step in the case of predicting cardiac arrest in a pediatric intensive care unit. Another issue brought to light by the second paper is that the individual steps required to use time series data for predictive modeling are more numerous and more complex than those used for modeling with traditional multivariate data. Even after complexities attributable to the design phase (addressed in our first paper) have been accounted for, the management and manipulation of the time series elements (the preprocessing steps in particular) are issues that are not present in a traditional multivariate modeling paradigm. In our methods, we present the issues that arise from the time series data elements: defining a reference time; imputing and reducing time series data in order to conform to a predefined structure that was specified during the design phase; and normalizing variable families rather than individual variable instances. The final manuscript, entitled: "Using Time-Series Analysis to Predict Cardiac Arrest in a Pediatric Intensive Care Unit" presents the results that were obtained by applying the theoretical construct and its associated methods (detailed in the first two papers) to the case of cardiac arrest prediction in a pediatric intensive care unit. Our results showed that utilizing the trend analysis from the time series data elements reduced the number of classification errors by 73%. The area under the Receiver Operating Characteristic curve increased from a baseline of 87% to 98% by including the trend analysis. In addition to the performance measures, we were also able to demonstrate that adding raw time series data elements without their associated trend analyses improved classification accuracy as compared to the baseline multivariate model, but diminished classification accuracy as compared to when just the trend analysis features were added (ie, without adding the raw time series data elements). We believe this phenomenon was largely attributable to overfitting, which is known to increase as the ratio of candidate features to class examples rises. Furthermore, although we employed several feature reduction strategies to counteract the overfitting problem, they failed to improve the performance beyond that which was achieved by exclusion of the raw time series elements. Finally, our data demonstrated that pulse oximetry and systolic blood pressure readings tend to start diminishing about 10-20 minutes before an arrest, whereas heart rates tend to diminish rapidly less than 5 minutes before an arrest.

PHENOTHIAZINES INDUCE ABORTIVE AUTOPHAGY LEADING TO CANCER CELL DEATH

Autophagy is an evolutionarily conserved process that functions to maintain homeostasis and provides energy during nutrient deprivation and environmental stresses for the survival of cells by delivering cytoplasmic contents to the lysosomes for recycling and energy generation. Dysregulation of this process has been linked to human diseases including immune disorders, neurodegenerative muscular diseases and cancer. Autophagy is a double edged sword in that it has both pro-survival and pro-death roles in cancer cells. Its cancer suppressive roles include the clearance of damaged organelles, which could otherwise lead to inflammation and therefore promote tumorigenesis. In its pro-survival role, autophagy allows cancer cells to overcome cytotoxic stresses generated the cancer environment or cancer treatments such as chemotherapy and evade cell death. A better understanding of how drugs that perturb autophagy affect cancer cell signaling is of critical importance toimprove the cancer treatment arsenal. In order to gain insights in the relationship between autophagy and drug treatments, we conducted a high-throughput drug screen to identify autophagy modulators. Our high-throughput screen utilized image based fluorescent microscopy for single cell analysis to identify chemical perturbants of the autophagic process. Phenothiazines emerged as the largest family of drugs that alter the autophagic process by increasing LC3-II punctae levels in different cancer cell lines. In addition, we observed multiple biological effects in cancer cells treated with phenothiazines. Those antitumorigenic effects include decreased cell migration, cell viability, and ATP production along with abortive autophagy. Our studies highlight the potential role of phenothiazines as agents for combinational therapy with other chemotherapeutic agents in the treatment of different cancers.

Structural and functional analysis of p84N5

Normal development and tissue homeostasis requires the carefully orchestrated balance between cell proliferation and cell death. Cell cycle checkpoints control the extent of cell proliferation. Cell death is coordinated through the activation of a cell suicide pathway that results in the morphologically recognizable form of death, apoptosis. Tumorigenesis requires that the balance between these two pathways be disrupted. The tumor suppressor protein Rb has not only been shown to be involved in the enforcement of cell cycle checkpoints, but has also been implicated in playing a role in the regulation of apoptosis. The manner in which Rb enforces cell cycle checkpoints has been well studied; however, its involvement in the regulation of apoptosis is still very unclear. p84N5 is a novel nuclear death domain containing protein that has been shown to interact with the N-terminus of Rb. The fact that it contains a death domain and the fact that it is nuclear localized possibly provides the first known mechanism for apoptotic signaling from the nucleus. The following study tested the hypothesis that the novel exclusively nuclear death domain containing protein p84N5 is an important mediator of programmed cell death and that its apoptotic function is reliant upon its nuclear localization and is regulated by unique functional domains within the p84N5 protein. We identified the p84N5 nuclear localization signal (NLS), eliminated it, and tested the functional significance of nuclear localization by using wild type and mutant sequences fused to EGFP-C1 (Clontech) to create wild type GFPN5 and subsequent mutants. The results of these assays demonstrated exclusive nuclear localization of GFPN5 is required for normal p84N5 induced apoptosis. We further conducted large-scale mutagenesis of the GFPN5 construct to identify a minimal region within p84N5 capable of interacting with Rb. We were able to identify a minimal sequence containing p84N5 amino acids 318 to 464 that was capable of interacting with Rb in co-immunoprecipitation assays. We continued by conducting a structural and functional analysis to identify the region or regions within p84N5 responsible for inducing apoptosis. Point mutations and small-scale deletions within the death domain of p84N5 lessened the effect but did not eliminate p84N5-induced cytotoxicity. Further analysis revealed that the minimal sequence of 318 to 464 of p84N5 was capable of inducing apoptosis to a similar degree as wild-type GFPN5 protein. Since amino acids 318 to 464 of p84N5 are capable of inducing apoptosis and interacting with Rb, we propose possible mechanisms whereby p84N5 may function in a Rb regulated manner. These results demonstrate that p84N5 induced apoptosis is reliant upon its nuclear localization and is regulated by unique functional domains within the p84N5 protein. ^