Recipient Organization
PURDUE UNIVERSITY
(N/A)
WEST LAFAYETTE,IN 47907
Performing Department
Veterinary Basic Medical Sciences
Non Technical Summary
Foodborne pathogens cause significant economic losses across all levels of industry and society, and a number of very dangerous pathogens are deadly. Even pathogens that cause no significant stress to normal healthy individuals can result in death for individuals who are immunocompromised, very young, or elderly. Thus, maintaining the integrity of the food chain is a critical part of our well-developed agricultural system. While there are many tools and technologies available for testing foods, reagents, liquids, etc. for pathogens, many tests are time consuming and/or expensive. Some can only be performed within high-technology environments, and some cannot be adapted to all types of samples.One of the important goals of this application is to integrate data management and create models for storage and access to data from different atomic spectra based instruments where data formats may be significantly different. While the USA is a highly technological society with tremendous infrastructure, many countries cannot afford a lot of current technology. Further, there is a significant growth of hand-held LIBS instruments that have the potential for expanding biologically driven testing such as food borne pathogen or toxin testing. Our goal is to to create a mechanism whereby data can be shared, but also the methods for collection, the nature of the instruments and the advantages and disadvantages can be readily identified. Since the vast majority of stored atomic spectra data is focused on chemistry, metallurgy, archeology and mining industries, expanding the biological domain can become a mechanism for improving the accuracy and relevance of data collected. This should create a useful tool for evaluation of LIBS based technologies in an independent and objective way.
Animal Health Component
30%
Research Effort Categories
Basic
40%
Applied
30%
Developmental
30%
Goals / Objectives
The overall goal of this project is to create tools that enhance our ability to classify microbial/toxin identities from food contamination. To achieve this, we must define new toolsets for analysis of data as the size and complexity of data have increased significantly. We have been developing technologies for example, that currently use the National Institute for Standards and Technology (NIST) data base (https://www.nist.gov/pml/atomic-spectra-database ) for providing analytical decisions as to the presence of different atomic elements. However, there are many problems with this approach as there are many differences in the assay design, the instrumentation and the base materials that impact the results. Further, there are huge gaps in the NIST database and an alternative approach that focuses on biological measurements and updating this to include common foods, food components would be a major boost to researches in this area. It would open up an opportunity for more studies using technologies like Laser Induced Breakdown Spectroscopy (LIBS) if more focused databases were available. A primary objective is to create a new database focused on the biological detection models we have been working on, including food borne pathogens, toxins, pesticides and a wide variety of chemical contaminants that might impact food and food projects. Creating more accurate classifiers for each pathogenic organism in a manner that allows for distribution to remote sites would give laboratories, food service centers, factories and agencies potential access to this database using the internet. Thespecific objectives of this project are: 1. Establish well defined spectra for organisms and toxins that comport with the emerging LIBS technologies 2. Develop visualization methods that allow a user to rapidly define abnormal biological samples 3. Establish a mechanism for cloud-based distribution of data and results 4. Create user-friendly interfaces for each of the output models defined. 5. Enhance the potential for phone-based download of data to/from the system.Specific targets for toxins will be: shiga toxins, botulinum and ricin. While other toxins are important, defining these will be very important. We will also target other molecules that have been shown to have toxic effect on individuals such as selenium.
Project Methods
1. Establish descriptive models for organisms and toxins that comport with the technologies available: Current LIBS databases are generic. Trying to define specific biological models is difficult and the variation is high. We propose to first create our own database focused on the biological molecule of interest to our group, and then identify other groups with similar datasets to add to the system.2. Develop visualization methods that allow a user to rapidly define abnormal samples: This requires a reasonable interface that will allow users to compare samples of similar nature. There are a variety of visualization methods available, but in many cases, these are not easily accessed. One of our objectives is to facilitate methods for integrating different types of data into a general display format. Such formats are currently used in evaluating web-based data, published literature, and other large data sets where the goal is to provide relational displays.3. Establish a mechanism for cloud-based distribution of data and results: This will likely be defined by current systems such as Amazon cloud networks which are quite economical and widely distributed.4. Create user-friendly interfaces for each of the output models defined. It is likely that we will have to define slightly different interfaces or systems that can convert data to a common format. These issues will become clear as we move toward this goal.5. Enhance the potential for phone-based download of data to the system. Clearly, with larger interest on hand-held, or portable data collection or distribution approaches, we much be consonant of a need to provide mechanisms for remote/portable device upload.