Kaizhi Tang

Predictive modeling of nanomaterial exposure effects in biological systems

Background Predictive modeling of the biological effects of nanomaterials is critical for industry and policymakers to assess the potential hazards resulting from the application of engineered nanomaterials. Methods We generated an experimental dataset on the toxic effects experienced by embryonic zebrafish due to exposure to nanomaterials. Several nanomaterials were studied, such as metal nanoparticles, dendrimer, metal oxide, and polymeric materials. The embryonic zebrafish metric (EZ Metric) was used as a screening-level measurement representative of adverse effects. Using the dataset, we developed a data mining approach to model the toxic endpoints and the overall biological impact of nanomaterials. Data mining techniques, such as numerical prediction, can assist analysts in developing risk assessment models for nanomaterials. Results We found several important attributes that contribute to the 24 hours post-fertilization (hpf) mortality, such as dosage concentration, shell composition, and surface charge. These findings concur with previous studies on nanomaterial toxicity using embryonic zebrafish. We conducted case studies on modeling the overall effect/impact of nanomaterials and the specific toxic endpoints such as mortality, delayed development, and morphological malformations. The results show that we can achieve high prediction accuracy for certain biological effects, such as 24 hpf mortality, 120 hpf mortality, and 120 hpf heart malformation. The results also show that the weighting scheme for individual biological effects has a significant influence on modeling the overall impact of nanomaterials. Sample prediction models can be found at http://neiminer.i-a-i.com/nei_models. Conclusion The EZ Metric-based data mining approach has been shown to have predictive power. The results provide valuable insights into the modeling and understanding of nanomaterial exposure effects.

Comparative hazard analysis and toxicological modeling of diverse nanomaterials using the embryonic zebrafish (EZ) metric of toxicity

Abstract The integration of rapid assays, large datasets, informatics, and modeling can overcome current barriers in understanding nanomaterial structure–toxicity relationships by providing a weight-of-the-evidence mechanism to generate hazard rankings for nanomaterials. Here, we present the use of a rapid, low-cost assay to perform screening-level toxicity evaluations of nanomaterials in vivo. Calculated EZ Metric scores, a combined measure of morbidity and mortality in developing embryonic zebrafish, were established at realistic exposure levels and used to develop a hazard ranking of diverse nanomaterial toxicity. Hazard ranking and clustering analysis of 68 diverse nanomaterials revealed distinct patterns of toxicity related to both the core composition and outermost surface chemistry of nanomaterials. The resulting clusters guided the development of a surface chemistry-based model of gold nanoparticle toxicity. Our findings suggest that risk assessments based on the size and core composition of nanomaterials alone may be wholly inappropriate, especially when considering complex engineered nanomaterials. Research should continue to focus on methodologies for determining nanomaterial hazard based on multiple sub-lethal responses following realistic, low-dose exposures, thus increasing the availability of quantitative measures of nanomaterial hazard to support the development of nanoparticle structure–activity relationships.

A text cube approach to human, social and cultural behavior in the twitter stream

Twitter is a microblogging website that has been useful as a source for human social behavioral analysis, such as political sentiment analysis, user influence, and spread of news. In this paper, we discuss a text cube approach to studying different kinds of human, social and cultural behavior (HSCB) embedded in the Twitter stream. Text cube is a new way to organize data (e.g., Twitter text) in multiple dimensions and multiple hierarchies for efficient information query and visualization. With the HSCB measures defined in a cube, users are able to view statistical reports and perform online analytical processing. Along with viewing and analyzing Twitter text using cubes and charts, we have also added the capability to display the contents of the cube on a heat map. The degree of opacity is directly proportional to the value of the behavioral, social or cultural measure. This kind of map allows the analyst to focus attention on hotspots of concern in a region of interest. In addition, the text cube architecture supports the development of data mining models using the data taken from cubes. We provide several case studies to illustrate the text cube approach, including public sentiment in a U.S. city and political sentiment in the Arab Spring.

NEIMiner: nanomaterial environmental impact data miner

As more engineered nanomaterials (eNM) are developed for a wide range of applications, it is crucial to minimize any unintended environmental impacts resulting from the application of eNM. To realize this vision, industry and policymakers must base risk management decisions on sound scientific information about the environmental fate of eNM, their availability to receptor organisms (eg, uptake), and any resultant biological effects (eg, toxicity). To address this critical need, we developed a model-driven, data mining system called NEIMiner, to study nanomaterial environmental impact (NEI). NEIMiner consists of four components: NEI modeling framework, data integration, data management and access, and model building. The NEI modeling framework defines the scope of NEI modeling and the strategy of integrating NEI models to form a layered, comprehensive predictability. The data integration layer brings together heterogeneous data sources related to NEI via automatic web services and web scraping technologies. The data management and access layer reuses and extends a popular content management system (CMS), Drupal, and consists of modules that model the complex data structure for NEI-related bibliography and characterization data. The model building layer provides an advanced analysis capability for NEI data. Together, these components provide significant value to the process of aggregating and analyzing large-scale distributed NEI data. A prototype of the NEIMiner system is available at http://neiminer.i-a-i.com/.

SocialCube: A Text Cube Framework for Analyzing Social Media Data

The recent development of social media (e.g., Twitter, Facebook, blogs, etc.) provides an unprecedented opportunity to study human social cultural behaviors. These data sources provide rich structured data (e.g., XML, relational tables, and categorical data) as well as unstructured data (e.g., texts). A significant challenge is to summarize and navigate structured data together with unstructured text data for efficient query and analysis. In this paper we introduce a text cube architecture designed to organize social media data in multiple dimensions and hierarchies for efficient information query and visualization from multiple perspectives. For example, an affective process cube allows the analyst to examine public reaction (e.g., sadness, anger) to a range of social phenomena. The text cube architecture also supports the development of prediction models using the summarized statistics stored in a data cube. For example, models that detect events, such as violent protests in the Egyptian Revolution, can be built using the linguistic features stored in an event data cube. These kinds of models represent higher level of knowledge representation and may help to develop more effective strategies for decision-making based on social media data.

Simulation anywhere any time: Web-based simulation implementation for evaluating order-to-delivery systems and processes

GM Enterprise Systems Laboratory (GMESL) has developed a standalone single user simulation program for evaluating and predicting order-to-delivery (OTD) systems and processes. In order for more people to be able to access this simulator, to share the simulation results, and to analyze simulation collaboratively, we have designed, developed and implemented an Internet-based three-tiered client/server framework, which consists of the three tiers: database, execution and user interface. The corresponding components are: database server, execution server, and Web based user interface. The relational database server enables users to interact with the persistent data sets for simulation study and maintains data integrity. The multi-agent based execution server guarantees stable user responsiveness by virtue of the multi-agent flexible architecture, accordingly achieving a high level of processing scalability. Finally the Web-based graphical user interface helps users to easily conduct the simulation study from anywhere at any time, and the visual simulation analysis tool helps users to make decisions effectively.

Information extraction from nanotoxicity related publications

High-quality experimental data are important when developing predictive models for studying nanomaterial environmental impact (NEI). Given that raw data from experimental laboratories and manufacturing workplaces are usually proprietary and small-scaled, extracting information from publications is an attractive alternative for collecting data. We developed an information extraction system that can extract useful information from full-text nanotoxicity related publications. This information extraction system consists of five components: raw data transformation into machine readable format, data preprocessing, ontology-based named entity recognition, rule-based numerical attribute extraction from both tables and unstructured text, and relation extraction among entities and attributes. The information extraction system is applied on a dataset made of 94 publications, and results in an acceptable accuracy. By storing extracted data into a table according to relations among the data, a dataset that can be used to predict nanomaterial environmental impact is obtained. Such a system is unique in current nanomaterial community, and can help nanomaterial scientists and practitioners quickly locate useful information they need without spending lots of time reading articles.

Cooperation in a multi-stage game for modeling distributed task delegation in a supply chain procurement problem

We develop an evolutionary method that combines reinforcement learning and fictitious playing to seek equilibrium solution for a multi-agent and multi-stage game in the context of supply chain procurement. The game is designed to model task delegation among a group of self-interested transportation companies which serve logistic shipment. The game involves more than two agents and multiple stages of matrix games. The integration of reinforcement learning and fictitious play overcomes the weaknesses of each approach and exploits their strengths. This innovative approach performs extraordinarily well on a game with three players, unknown number of stages, and large gaps of payoff values.

A user centered approach to developing information visualization module for NEIMiner

NEIMiner is an integrated information system for studying the nanomaterial environmental impact (NEI). However, there is a lack of visual analytic tools that efficiently query and present large-scaled bibliography meta-data and NEI characterizations in a meaningful way. This paper presents the design and implementation efforts of developing the information visualization (InfoVis) module for NEIMiner. We first describe a user centered design approach to identify the analysis tasks, to select suitable visual representations, and to iteratively validate and improve the development. We then show that how existing techniques, such as graph simplification, enriched visualization algorithms and interactive features, can be usefully combined to aid users gaining insights. We demonstrate the utility of Info Vis through scenarios of constructing co-authorship network, bibliography keywords network, and nanomaterial terms co-occurrence network. We implement our techniques as a Drupal module. Our design is supportive for analysts and researchers to identify concepts and relationships in studying environmental impact of nanomaterial.

Predictive modeling of nanomaterial biological effects

Nanomaterial environmental impact (NEI) modeling is critical for industry and policymakers to assess the unintended biological effects (e.g. mortality, malformation, growth inhibition) resulting from the application of engineered nanomaterials. The scope of NEI modeling covers nanomaterial physical, chemical and manufacturing properties, exposure and study scenarios, environmental and ecosystem responses, biological responses, and their interactions. In this paper, we introduce a data mining approach to modeling the biological effects of nanomaterials. Data mining techniques can assist analysts in developing risk assessment models for nanomaterials. Using an experimental dataset on the toxicity of nanomaterials to embryonic zebrafish, we conducted case studies on modeling the overall effect/impact of nanomaterials and the specific toxic end-points such as mortality, delayed development, and morpholigcal malformations and behavioral abnormalities. The results show that different biological effects have different modeling accuracy given the same set of algorithms and data. The results also show that the weighting scheme for different biological effects has a significant influence on modeling the overall biological effect. These results provide insights into the understanding and modeling of nanomaterial biological effects.