My primary interests are in statistical methods for biomedical research. I am also interested in educational research from a sociological perspective. My current projects include the following.

  • GENOMICS is the large-scale study of genes, including their DNA and RNA sequences, expression, function and interactions with other genes. Microarray technology allows for high-throughput data measurements on numerous genetic features, such as mRNA, miRNA, methylation, SNPs, copy number and protein expression. Bioinformatics applies “informatic” disciplines such as mathematics, computer science and statistics to organize, understand and analyze such molecular data from microarray platforms. As a statistician I am interested in understanding these data, how they are generated, their variation, and ultimately what they tell us about the development of abnormal processes such as run-away cell growth that leads to cancer. Understanding  molecular differences between normal and disease subjects can lead to a better understanding of a disease or new treatments. My work in bioinformatics is largely motivated by collaborations with cancer  investigators at Baylor College of Medicine, Texas Children’s Hospital and M.D. Anderson Cancer Center. Some of my work includes:
    1. Osterosarcoma is the most malignant bone tumor in children and young adults and relatively little is known about its genomic signatures.  This project is with Dr. Ching Lau (BCM-TCH) whose lab aims to characterize of osteosarcoma genome via bioinformatic methods. Here we work to understand both the individual and integrated roles of mRNA, miRNA, methylation, and copy number in osteosarcoma.
    2. Another project on pediatric osteosarcoma is with Dr. Chris Man (BCM-TCH). Here we work with cell lines to identify biomarkers for osteosarcoma, as well as to understand the role of some specific proteins associated with osteosarcoma.
    3. Genomic characterization of cervical cancer. This is joint work with department colleagues Dennis Cox and Neely Atkinson.
    4. The role of stem cells in cancer with Dr. Sendurai Mani (MDACC). This work concerns how stem cells arise in cancer and how they may be involved in metastasis. Much of this work involves mRNA and miRNA.
  • PROTEOMICS is the large-scale study of proteins, including their abundance, interactions, structure, and function. Most bioinformatic efforts have been in genomics, but ultimately it is proteins that affect cellular processes that determine diseases such as cancer. Proteins also allow us to identify biomarkers or biosignatures for various disease processes. My biomedical research effort in proteomics is currently in the following areas:
    1. Protein-protein interactions. In this study with Drs. Junjie Chen and Xu Li we use spectral counts from mass spectrometry signals to infer whether pairs of proteins interact with each other under certain cellular conditions.  The ultimate aim is establish a database of protein interactions for cancer research. The statistical problem concerns developing a method that predicts the probability of two proteins interacting while minimizing false-positive and false-negative rates. Ben White, a Ph.D. student in statistics is contributing to this project.
    2. With Drs. Tim Palzkill and Rose Mikulski of BCM we investigate proteins that allow bacteria to resist to a wide group of antibiotics. Resistance to antibiotics is a growing problem in medical practice and understanding the sequences of such proteins may allow combat the bacterial resistance. Here we use statistical methods to identify mutant sequences that confer the bacterial resistance.
    3. In the pediatric osteosarcoma project with Dr. Chris Man we also investigate p27, a well-known tumor suppressor gene that regulates normal cell cycle progression when it is located in the nucleus, but for unknown reasons develops an oncongenic function when located in the cytoplasm. This is the so-called mislocalization of p57.  In this project we use a panel of targeted genomic and proteomic methods to dissect the mechanism of the cytoplasmic p27 in the promotion of metastasis in osteosarcoma and other cancers. One goal is to test if cytoplasmic p27 can be used as a novel therapeutic target to abolish metastasis in cancer.
  • MEDICAL IMAGING. I also have an interest in the area of medical imaging and have published several articles with Drs. Thomas Guerrero and Edward Castillo of MDACC and Dr. Richard Castillo of UTMB. My current work continues with Richard Castillo in the use of deformable image registration (DIR) in medical image analysis and functional lung imaging from computed tomography (CT). DIR has many potential applications in diagnostic imaging and radiotherapy. As a developing technology part of our effort involves investigation of statistical methods for evaluation of DIR spatial accuracy for purposes of validation and acceptance testing, as well as quality assurance in routine clinical application. We also investigate statistical methods for utilizing DIR output to extract functional information.
    1. Educational research. Over the past year (since fall 2013) I have been working with Dr. Ruth Lopez-Turley of the sociology department on education research pertaining to the Houston Independent School District (HISD). The work involves analyses of various studies and experiments aimed at understanding socioeconomic gaps in achievement and attainment by the students in HISD. This is interesting work from a statistical perspective as it involves a number of  methods not normally found in the tool bag of many statisticians working in the sciences. Factor analysis, latent variable modeling, instrumental variables, and  two-stage least squares are standard methods for this type of work.
    2. I also have recently started a collaboration with Dr. Bridget Gorman of the sociology department, but in the direction of Mexican migration research. Presently (summer 2014), we are conducting a survey in a small Mexican town to study the implications of unauthorized migration on health and medical care.

Comments are closed.