• Multi-modality Informatics Approach for Early Stage Cancer Diagnosis

    Breast cancer is the most commonly diagnosed cancer in women, and is the second leading cause of cancer death among women in U.S. However, current screening methods for early detection of breast cancer often lead to over-diagnosis and subsequent over-treatment, and recent studies estimated that breast cancer was over-diagnosed in 1.3 million U.S. women in the past 30 years.

  • Develop new mouse models for human cancer

    Sporadic tumors, which account for the majority of all human cancers, evolve as the result of a step-wise accumulation of genetic alterations resulting in uncontrolled cell proliferation and a lack of response to apoptotic cues. Such genetic alterations include point mutations, deletions, duplication/amplification, and translocations and these alterations can lead to the enhanced or decreased activity of the expressed protein. These alterations are referred to as ‘gain-of-function’ or ‘loss-of-function’ mutations, respectively. The affected genes are termed oncogenes or tumor suppressors, respectively. Within the last decade, the availability of a complete sequence-based map of the human genome, coupled with significant technological advances, has revolutionized the search for somatic alterations in tumor genomes. Within a given tumor type there are many infrequently mutated genes and a few frequently mutated genes, resulting in incredible genetic heterogeneity. The resulting catalogues of somatic alterations will point to candidate cancer genes, but requiring further validation to determine whether they have a causal role in tumorigenesis. The availability of gene targeting and transgenic technology in the mouse gives us unparalleled opportunities to test the functional significance of genetic changes in tumor development. Another one of the broad and long-term goals of my laboratory is to develop new mouse models for human cancer. These mouse models not only will increase our understanding of genetic aberration associated with cancer progression, but also will potentially help to identify personalized medicine for cancer patients, which may eventually contribute to a decrease in morbidity and mortality of cancer.

  • A systems biology approach to identification of genetic networks controlling susceptibility to cancer risk

    Genetic susceptibility plays a role in many types of cancer. Identifying the genes involved in susceptibility to cancer may have potential utility in risk management, lead to greater understanding of the biological pathways involved in cancer development, and elucidate how environmental factors exert their effects in combination with genetic variants. Major problems in assigning biomarkers of human cancer risk are that humans are genetically heterogeneous and exposures are pervasive and difficult to quantitate. Parallel exposure to multiple chemicals and lifestyle factors that can negatively impact health further confound efforts to assign risk. Mouse models offer many advantages for the study of the genetic basis of complex traits, including cancer, because of our ability to control both the genetic and environmental components of risk. The goal is the understanding of all stages of multi-step carcinogenesis in the mouse, in particular the relationships between germ line predisposition and somatic genetic changes in tumors. The identification of human homologues of these predisposition genes and discovery of their roles in carcinogenesis will ultimately be important for the development of methods for prediction of risk, diagnosis, prevention and therapy for human cancers. We will exploit the variation in susceptibility to radiation-induced cancers between mouse strains to identify the combinations of quantitative trait loci (QTLs) that control the radiation response. The power of classical mouse genetics will be complemented by new approaches involving haplotyping to refine the genomic locations of QTLs, together with sophisticated genetic analysis of the somatic events in radiation-induced cancers using newly developed high throughput genome wide BAC microarrays. The relationship between somatic events and germline polymorphism that influence risk will be investigated by analysis of allele-specific genetic alterations in tumors that occur within genomic regions containing tumor susceptibility genes. Gene expression microarray technology will be used to identify candidate genes and pathways implicated in radiation-induced acute responses and tumorigenesis in vivo. Expression array analysis will be carried out on normal and tumor tissues from mice that are sensitive or resistant to radiation-induced tumorigenesis, to look for genes that may be differently expressed due to polymorphisms in gene promoter or controlling regions, or in coding regions of upstream regulatory genes. This comprehensive systems biology approach may identify specific genes or pathways that are differentially controlled between mouse strains, and contribute to variation in susceptibility to radiation-induced carcinogenesis.

  • Big data oriented imaging bio-markers identification towards personalized therapy

    Adult primary brain tumors such as gliomas are characterized by enormous cellular diversity captured by grading. For gliomas, tumor grade is based on the region with the highest level of aberrant histopathology. Recently, a large number of subtypes have been characterized based on morphological variants and their molecular characterization showing an enormous heterogeneity that is only being discovered. This poses an enormous challenge in interpreting these subtypes and understanding their clinical and molecular associations.This project aims to develop open source advanced image-based modeling algorithms and software and to couple them with a bioinformatics system for the analysis of brain tumors. The net results are: a more robust identification of tumor subtypes; hypothesis generation for the molecular basis of each subtype; creation of a publicly available databank where new tissue sections can be compared against an existing database of prognostic and predictive subtypes and their molecular signatures; and potentially enabling new opportunities for personalized therapy.