Since 2014, I have had mentored 15 undergraduate students in various research projects in interdisciplinary areas including bioanalytical chemistry, environmental chemistry, medicinal chemistry, and chemometrics. These students have presented their research in local and international conferences, published manuscripts, and won external travel fellowships. My present research projects are funded by two NSF grants via the Louisiana Board of Regents. In 2015, I co-founded a company called “LipidTech, LLC” (http://lipidtech.weebly.com), an off shoot of my PhD research involving the development of a spectrophotometric and chemometric method for the simultaneous determination of 7 polyunsaturated fatty acids in human serum and biological samples.
I. Rapid authentication and detection of commercial honey samples
Honey is the third most adulterated product in the world, and companies less interested in honey authenticity and quality control are able to sell their products at a lower rate. Honey adulteration is a serious global issue that has significant economic and organoleptic consequences. In the recent years, the addition of adulterants such as corn, cane, beet, and rice syrup to honey have become very prevalent and hard to detect. Current methods for detecting such adulterants are typically complex, expensive, and time consuming. The overall goal of this study is to develop an accurate, robust, and facile analytical method for the simultaneous quantification of honey adulterants consisting of corn, cane, beet, and rice syrup in honey using chemometrics and a multi-sensor data fusion technique. Multi-sensor data fusion technique combines the synergistic effects of the information obtained from multiple process analyzers including Fourier Transform Infrared (FTIR) and nuclear magnetic resonance (NMR) spectroscopy as in this study, to simultaneously detect various adulterants in honey. Specifically, this study aims to: 1) develop a method for the simultaneous quantification of honey adulterants by chemometrics and a data fusion technique, and 2) determine the authenticity and floral origin of various honey samples. The innovative application of various spectroscopic, chemometric, and ensemble learning techniques in this study can potentially improve overall honey quality, sensorial characteristics, and safety. In order to accomplish this project, honey samples will be obtained from my collaborators from the USDA in Baton Rouge, Louisiana, Adee Honey Farms, and several hobbyists. Adee Honey Farms can readily provide honey samples from SD, NE, TX, LA, MS, UT, and CA.
II. Introducing Chemometrics to the Undergraduate Laboratory Curriculum: An Investigation into Mixture Analysis Using Food Coloring Dyes
Chemometrics is a relatively new area of science and a branch of analytical chemistry that involves the use of statistics and computer applications in analytical chemistry. Over the years, this particular field has not received a wide span of attention in the undergraduate curriculum due to the daunting mathematical calculations that are involved in the subject. In this study, an introduction to chemometrics techniques will be performed in the undergraduate laboratory curriculum by utilizing a synthetic five-component mixture of food coloring dyes. Specifically, I will introduce the use of the commonly used chemometric techniques, including classical least squares (CLS), partial least squares (PLS) and principal component regression (PCR) using the UV-VIS and FTIR spectra. Students will also be introduced to the concept of principal component analysis (PCA) and clustering techniques using the sample mixtures generated from the design of experiments. Finally, food samples with unknown colored dyes will be investigated using the developed calibration model. This exercise can be completed in one four-hour laboratory.
III. Prediction of cardiorespiratory fitness gains via chemometric approaches
Cardiorespiratory fitness (CRF) is defined as the ability of the cardiovascular and respiratory systems to supply oxygen to the working muscle during sustained exercise. CRF is measured at maximal or submaximal workloads. CRF is a strong predictor of the risk of premature death in adults. This proposal will use the resources of the NIH-funded HERITAGE Family Study from the Pennington Biomedical Research Center (PBRC) in Baton Rouge, LA in collaboration with my colleague, Dr. Claude Bouchard (https://en.wikipedia.org/wiki/Claude_Bouchard). The HERITAGE Family Study (HEalth, RIsk factors, exercise Training And GEnetics) is funded by NIH for 20 years and consists of a wide array of human biological samples ranging from muscle, urine, feces, blood, etc of 450 European Americans and 200 African Africans. My research group has been given access to all these samples. Further, readily available data from PBRC will be used for two aims:
AIM 1: Prediction of CRF gains based on behavioral, physiological, metabolic, genetic and metabolomics factors. The development of univariate, multivariate and holistic predictive models will allow to identify what are the most substantive predictors of CRF response to a standardized exercise dose as was the case in HERITAGE.
AIM 2: Implementation of a data-driven approach utilizing chemometric techniques to shed light on the intricate network of behavioral and biological interactions impacting CRF response to a dose of exercise.
IV. Simultaneous rapid testing of microbial pathogens in pharmaceutical products using Fourier transform infrared (FTIR) spectroscopy and chemometrics
Pharmaceutical products are subject to microbial contamination that results to changes in their physico-chemical properties and subsequent product degradation. This problem costs pharmaceutical companies significant financial loss annually through drug contamination and production stoppage. In general, traditional microbiological methods in pharmaceutical industries tend to be labor-intensive and time-consuming. These has resorted to some pharmaceutical companies to outsource microbiological tests to third party laboratories. We are proposing the development of a rapid test method of four microbial pathogens (Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, and Candida albicans) in pharmaceutical products using Fourier Transform Infrared (FTIR) spectroscopy and chemometrics. These pathogens were chosen because they are the most commonly found microbes in pharmaceutical products and are the required microbial tests for nonsterile products. The specific objectives, therefore, would include 1) The development of a chemometric model for the identification and quantification of microbial pathogens, 2) Validation of the accuracy of the chemometric model, and 3) Application of the model in pharmaceutical products for microbe quantification and identification.
I. Rapid authentication and detection of commercial honey samples
Honey is the third most adulterated product in the world, and companies less interested in honey authenticity and quality control are able to sell their products at a lower rate. Honey adulteration is a serious global issue that has significant economic and organoleptic consequences. In the recent years, the addition of adulterants such as corn, cane, beet, and rice syrup to honey have become very prevalent and hard to detect. Current methods for detecting such adulterants are typically complex, expensive, and time consuming. The overall goal of this study is to develop an accurate, robust, and facile analytical method for the simultaneous quantification of honey adulterants consisting of corn, cane, beet, and rice syrup in honey using chemometrics and a multi-sensor data fusion technique. Multi-sensor data fusion technique combines the synergistic effects of the information obtained from multiple process analyzers including Fourier Transform Infrared (FTIR) and nuclear magnetic resonance (NMR) spectroscopy as in this study, to simultaneously detect various adulterants in honey. Specifically, this study aims to: 1) develop a method for the simultaneous quantification of honey adulterants by chemometrics and a data fusion technique, and 2) determine the authenticity and floral origin of various honey samples. The innovative application of various spectroscopic, chemometric, and ensemble learning techniques in this study can potentially improve overall honey quality, sensorial characteristics, and safety. In order to accomplish this project, honey samples will be obtained from my collaborators from the USDA in Baton Rouge, Louisiana, Adee Honey Farms, and several hobbyists. Adee Honey Farms can readily provide honey samples from SD, NE, TX, LA, MS, UT, and CA.
II. Introducing Chemometrics to the Undergraduate Laboratory Curriculum: An Investigation into Mixture Analysis Using Food Coloring Dyes
Chemometrics is a relatively new area of science and a branch of analytical chemistry that involves the use of statistics and computer applications in analytical chemistry. Over the years, this particular field has not received a wide span of attention in the undergraduate curriculum due to the daunting mathematical calculations that are involved in the subject. In this study, an introduction to chemometrics techniques will be performed in the undergraduate laboratory curriculum by utilizing a synthetic five-component mixture of food coloring dyes. Specifically, I will introduce the use of the commonly used chemometric techniques, including classical least squares (CLS), partial least squares (PLS) and principal component regression (PCR) using the UV-VIS and FTIR spectra. Students will also be introduced to the concept of principal component analysis (PCA) and clustering techniques using the sample mixtures generated from the design of experiments. Finally, food samples with unknown colored dyes will be investigated using the developed calibration model. This exercise can be completed in one four-hour laboratory.
III. Prediction of cardiorespiratory fitness gains via chemometric approaches
Cardiorespiratory fitness (CRF) is defined as the ability of the cardiovascular and respiratory systems to supply oxygen to the working muscle during sustained exercise. CRF is measured at maximal or submaximal workloads. CRF is a strong predictor of the risk of premature death in adults. This proposal will use the resources of the NIH-funded HERITAGE Family Study from the Pennington Biomedical Research Center (PBRC) in Baton Rouge, LA in collaboration with my colleague, Dr. Claude Bouchard (https://en.wikipedia.org/wiki/Claude_Bouchard). The HERITAGE Family Study (HEalth, RIsk factors, exercise Training And GEnetics) is funded by NIH for 20 years and consists of a wide array of human biological samples ranging from muscle, urine, feces, blood, etc of 450 European Americans and 200 African Africans. My research group has been given access to all these samples. Further, readily available data from PBRC will be used for two aims:
AIM 1: Prediction of CRF gains based on behavioral, physiological, metabolic, genetic and metabolomics factors. The development of univariate, multivariate and holistic predictive models will allow to identify what are the most substantive predictors of CRF response to a standardized exercise dose as was the case in HERITAGE.
AIM 2: Implementation of a data-driven approach utilizing chemometric techniques to shed light on the intricate network of behavioral and biological interactions impacting CRF response to a dose of exercise.
IV. Simultaneous rapid testing of microbial pathogens in pharmaceutical products using Fourier transform infrared (FTIR) spectroscopy and chemometrics
Pharmaceutical products are subject to microbial contamination that results to changes in their physico-chemical properties and subsequent product degradation. This problem costs pharmaceutical companies significant financial loss annually through drug contamination and production stoppage. In general, traditional microbiological methods in pharmaceutical industries tend to be labor-intensive and time-consuming. These has resorted to some pharmaceutical companies to outsource microbiological tests to third party laboratories. We are proposing the development of a rapid test method of four microbial pathogens (Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, and Candida albicans) in pharmaceutical products using Fourier Transform Infrared (FTIR) spectroscopy and chemometrics. These pathogens were chosen because they are the most commonly found microbes in pharmaceutical products and are the required microbial tests for nonsterile products. The specific objectives, therefore, would include 1) The development of a chemometric model for the identification and quantification of microbial pathogens, 2) Validation of the accuracy of the chemometric model, and 3) Application of the model in pharmaceutical products for microbe quantification and identification.