Presenter
Paul Scheet
Biography
Paul Scheet, Ph.D., is Associate Professor at The University of Texas MD Anderson Cancer Center in the Dept. of Epidemiology with joint appointments in the Depts. of Genomic Medicine and Translational Molecular Pathology. A statistical geneticist with interests in complex disease and cancer genomics, Dr. Scheet serves as Leader of MD Anderson’s CCSG “Risk, Detection and Outcomes” Program, and since 2016 as Special Assistant to the Chair ad interim.
He leads or co-leads projects to discover risk alleles for pancreatic cancer and study the evolution to metastasis from primary lung cancer. Of particular interest in recent years is the detection of acquired chromosomal alterations existing at low intra-sample frequencies, such as when a small proportion of the cells in a heterogeneous mixture exhibit these mutations. Dr. Scheet’s lab has helped pioneer methods to more accurately distinguish these alterations using information on the order of inherited alleles on a chromosome, i.e. the haplotypes. This technique, implemented in the software hapLOH has applications in cancer genomics, early detection of cancer, and profiling of normal or pre-disease tissues exposed to carcinogens or aging. Profiles of nonmalignant tissues (premalignant, field cancerization, or putatively healthy) may help us better understand critical molecular changes that occur early in the transformation to disease or malignancy and leverage these mutations as biomarkers for cancer risk, progression, or outcomes.
After completing a B.A. in Biology, Dr. Scheet worked on the Human Genome Project at the Genome Center at Washington University in St. Louis, performing Sanger sequencing and informatics. To complement interests in large-scale genetic data and population genetics, he pursued studies in Statistics, with a master’s (Iowa, 2000) and Ph.D. (Washington, 2006), followed by a postdoctoral fellowship at Univ. of Michigan, before joining MD Anderson in 2008. At Washington, he developed a statistical model that captured features of the genealogy of chromosomes in a population. This model approximated important features of the established "coalescent" model but allowed for computationally efficient inference of important features, such as haplotypes and missing genotypes from panels of reference data. The software for this model, fastPHASE, ushered in a host of similar methods that have become part of standard practice in the analysis genome-wide association studies (GWAS).
A member of The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences since 2008, Dr. Scheet has mentored several students to completion of their doctorate degrees and serves on numerous Ph.D. committees at the GSBS and at other institutions in the Texas Medical Center.
He leads or co-leads projects to discover risk alleles for pancreatic cancer and study the evolution to metastasis from primary lung cancer. Of particular interest in recent years is the detection of acquired chromosomal alterations existing at low intra-sample frequencies, such as when a small proportion of the cells in a heterogeneous mixture exhibit these mutations. Dr. Scheet’s lab has helped pioneer methods to more accurately distinguish these alterations using information on the order of inherited alleles on a chromosome, i.e. the haplotypes. This technique, implemented in the software hapLOH has applications in cancer genomics, early detection of cancer, and profiling of normal or pre-disease tissues exposed to carcinogens or aging. Profiles of nonmalignant tissues (premalignant, field cancerization, or putatively healthy) may help us better understand critical molecular changes that occur early in the transformation to disease or malignancy and leverage these mutations as biomarkers for cancer risk, progression, or outcomes.
After completing a B.A. in Biology, Dr. Scheet worked on the Human Genome Project at the Genome Center at Washington University in St. Louis, performing Sanger sequencing and informatics. To complement interests in large-scale genetic data and population genetics, he pursued studies in Statistics, with a master’s (Iowa, 2000) and Ph.D. (Washington, 2006), followed by a postdoctoral fellowship at Univ. of Michigan, before joining MD Anderson in 2008. At Washington, he developed a statistical model that captured features of the genealogy of chromosomes in a population. This model approximated important features of the established "coalescent" model but allowed for computationally efficient inference of important features, such as haplotypes and missing genotypes from panels of reference data. The software for this model, fastPHASE, ushered in a host of similar methods that have become part of standard practice in the analysis genome-wide association studies (GWAS).
A member of The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences since 2008, Dr. Scheet has mentored several students to completion of their doctorate degrees and serves on numerous Ph.D. committees at the GSBS and at other institutions in the Texas Medical Center.
Presentations
Workshop
Applications
Deep Learning
Exascale
W