Statistics Webinars

Organizers (in alphabetic order): Prof. Hongshik Ahn (), Prof. Stephen Finch (), Prof. Pei-fen Kuan (), Prof. Song Wu (), Prof. Haipeng Xing (), Prof. Wei Zhu ()

Zoom Link and Password: Request by e-mailing to Mrs. Laurie Dalessio (laurie.dalessio@stonybrook.edu)

Spring 2021 Semester

Prof. Chao Chen | 12 March 2021 | Learning with Topological Information - Image Analysis and Label Noise

Speaker: Chao Chen

Organization: Stony Brook University, Department of Biomedical Informatics

Title: Learning with Topological Information - Image Analysis and Label Noise

Date/Time: 12 Mar. 2021, Friday, 2:00 - 3:00 PM EST

Abstract: Modern machine learning faces new challenges. We are analyzing highly complex data with unknown noise. Topology provides novel structural information to model such data and noise. In this talk, we discuss two directions in which we are using topological information in the learning context. In image analysis, we propose a topological loss to segment and to generate images with not only per-pixel accuracy, but also topological accuracy. This is necessary in analysis of images of fine-scale biomedical structures such as neurons, vessels, etc. Extracting these structures with correct topology is essential for the success of downstream analysis. Meanwhile, we discuss how to use topological information to train classifiers robust to label noise. This is important in practice especially when we are using deep neural networks which tend to overfit noise. These results have been published in NeurIPS, ECCV, ICML and ICLR.

Prof. Xuefeng Wang | 2 April 2021 | New methods for discovering cancer disparity and immune related biomarkers

Speaker: Xuefeng Wang

Organization: H. Lee Moffitt Cancer Center & Research Institute

Title: New methods for discovering cancer disparity and immune related biomarkers

Date/Time: 2 Apr. 2021, Friday, 2:00 - 3:00 PM EST

Abstract: Modern machine learning methods have been successfully applied in many areas of biomedical research, and they have emerged as promising tools for biomarker discovery with genomic data. However, there is a lack of systematic research on approaches and tools in unraveling cancer immune and disparity related mechanisms, which are the two most disruptive topics in the field of cancer research nowadays. In this talk, I will share our recent work in addressing some of the analytical challenges, and discuss further solutions that can pull insights and deliver translational signatures from highly heterogenous cancer population datasets. First, I will discuss how a set of efficient machine learning approaches (including kernel machine and gradient boosting) can benefit the prognostic biomarker search by analyzing high-dimensional genomics data. In the second part of the talk, a new scheme for characterizing immune cell compositions from bulk and single-cell tumor gene expression will be introduced. And last, but not least, I will present our most recent work and findings on dissecting racial and socioeconomic disparities based upon a large-scale clinical oncology dataset.

Prof. Hongyu Zhao | 7 May 2021 | Predicting Disease Risk from Genomics Data

Speaker: Hongyu Zhao

Organization: Yale University, Department of Biostatistics

Title: Predicting Disease Risk from Genomics Data

Date/Time: 7 May. 2021, Friday, 1:45 - 2:45 PM EST

Abstract: Accurate disease risk prediction based on genetic and other factors can lead to more effective disease screening, prevention, and treatment strategies. Despite the identifications of thousands of disease-associated genetic variants through genome-wide association studies in the past 15 years, performance of genetic risk prediction remains moderate or poor for most diseases, which is largely due to the challenges in both identifying all the functionally relevant variants and accurately estimating their effect sizes. Moreover, as most genetic studies have been conducted in individuals of European ancestry, it is even more challenging to develop accurate prediction models in other populations. Furthermore, many studies only provide summary statistics instead of individual level genotype and phenotype data. In this presentation, we will discuss a number of statistical methods that have been developed to address these issues through jointly estimating effect sizes (both across genetic markers and across populations), modeling marker dependency, incorporating functional annotations, and leveraging genetic correlations among different diseases. We will demonstrate the utilities of these methods through their applications to a number of complex diseases/traits in large population cohorts, e.g. the UK Biobank data. This is joint work with Wei Jiang, Yiming Hu, Yixuan Ye, Geyu Zhou, Qiongshi Lu, and others.

TBD | 4 June 2021 | TBD

Speaker: TBD

Title: TBD`

Date/Time: 4 June. 2021, Friday, TBD

Abstract: TBD