Posts classified under: Workshops

BIODS 260
10/13/22
1:30 pm-2:30 pm
Lynn Petukhova
Assistant Professor, Epidemiology and Dermatology at the Columbia University Medical Center

Title: Leveraging Information in the Human Genome to Improve Skin Health and to Advance the Practice of Dermatology

Abstract: The process of diagnosing a patient historically has largely relied on clinical observations of symptoms by physicians. Limitations of a clinical diagnosis have been identified with the use of genetic and genomic technologies, which demonstrate that a molecular diagnosis derived from biomedical data can provide greater diagnostic accuracy and inform subsequent management. I conduct human genetic studies as a starting point for leveraging information in the human genome to improve the accuracy and utility of a skin disease diagnosis. Statistical evidence for an association between an inherited genetic variant and a disease outcome is a definitive marker for a disease mechanism, but does not provide adequate resolution of the mechanism for clinical translation. The scale and complexity of biomedical data that is available to define disease mechanisms requires data-driven approaches to identify salient features and to detect patterns among them that link disease mechanisms to interventions and outcomes. Using the hair follicle as a model organ to understand mappings between disease mechanisms and clinical diagnoses, our group is using clustering, network, and tensor factorization methods to discover clinically relevant relationships among genetically-derived disease entities. I will present results from three studies that our group is conducting that leverages knowledge about inherited genetic variants, disease genes, pathways, and/or comorbidities to define an underlying causal structure of skin disease pathogenesis and to identify key genetic regulators of hair follicle health.

Suggested Readings:

https://www.nature.com/articles/s41598-017-16050-9

https://dl.acm.org/doi/abs/10.1145/3368555.3384464

Zoom link: https://stanford.zoom.us/j/92865685887pwd=YjlUM1cxOHZ4UnBZMkhqcG1JYzFNdz09

Meeting ID: 928 6568 5887

Password: 219826

PDF Flier

BIODS 260 presents:
Dr. Lee Hood
10/20/22
1:30 pm-2:50 pm

Dr. Leroy Hood, MD, PhD, CEO/Founder of Phenome Health and Chief Strategy Officer and Professor of the Institute of Systems Biology

Suggested readings:

• A wellness study of 108 individuals using personal, dense, dynamic data clouds; 2017; ND Price et al; Nature Biotechnology
• Multi-Omics Resolves a Sharp Disease-State Shift between Mild and Moderate COVID-19; 2020; Y Su et al; Cell
• Multiomic blood correlates of genetic risk identify presymptomatic disease alterations; 2020; M Wainberg et al; Proceedings of the National Academy of Sciences 
• Untargeted longitudinal analysis of a wellness cohort identifies markers of metastatic cancer years prior to diagnosis; 2020; AT Magis et al; Scientific Reports

Zoom Link:

https://stanford.zoom.us/j/92874055477pwd=aThzNmpmNEQ1L2FjV0E5ZXF5SDR1UT09&from=addon

PDF Flier

BIODS 260 presents:
Dr. Adrian Buganza-Tepole
10/27/22
130 pm-2:30 pm
Location: MSOB x303

Dr. Adrian Buganza-Tepole is an Associate Professor of Mechanical Engineering and Biomedical Engineering (courtesy) at Purdue University. He obtained his Ph.D. in Mechanical Engineering from Stanford University in 2015 and was a postdoctoral fellow at Harvard University for a year before joining Purdue as a faculty member in 2016. He was also a Miller Visiting Professor at UC Berkeley during Spring 2022. His group studies the interplay between mechanics and mechanobiology of skin. Using computational simulation, machine learning, and experimentation, his group seeks to characterize the multi-scale mechanics of skin to understand the fundamental mechanisms of this tissue’s mechano-adaptation in order to improve clinical diagnostics and interventional tools.

Title: Data-driven skin biophysics  

Abstract: 

The recent explosion in machine learning (ML) and artificial intelligence (AI) algorithms has started a revolution in many engineering fields, including computational biophysics. This talk focuses on our recent efforts to leverage ML methods to increase our fundamental understanding of skin and its unique ability to adapt to mechanical cues. The first project that will be described is skin growth in tissue expansion, a popular reconstructive surgery technique that grows new skin in response to sustained supra-physiological loading. We have created computational models that combine mechanics and mechanobiology to describe the deformation and growth of expanded skin. Together with experiments on a porcine model, and leveraging ML tools such as multi-fidelity Gaussian processes, we have performed Bayesian inference to learn mechanistically how skin grows in response to stretch. The second half of the talk will explore how mechanical cues can be key drivers of wound healing pathologies such as fibrosis. I will show computational models of reconstructive surgery and wound healing for a murine model of wound healing and in patient-specific cases. Once again, ML methods enable new kinds of analyses such as optimization under uncertainty and inverse parameter calibration which are not achieved with traditional approaches.

Suggested readings: 

Han T, et al. Bayesian calibration of a computational model of tissue expansion based on a porcine animal model. Acta Biomaterialia. 2022;137:136-46.

Tac V, Costabal FS, Tepole AB. Data-driven tissue mechanics with polyconvex neural ordinary differential equations. Comput Method Appl Mech Eng. 2022;398:115248.

Sohutskay DO, Tepole AB, Voytik-Harbin SL. Mechanobiological wound model for improved design and evaluation of collagen dermal replacement scaffolds. Acta Biomaterialia. 2021;135:368-82.

Lee T, Bilionis I, Tepole AB. Propagation of uncertainty in the mechanical and biological response of growing tissues using multi-fidelity Gaussian process regression. Comput Method Appl Mech Eng. 2020;359:112724.

Zoom info:

• • Meeting ID: 937 0175 0367
  • Password: 249596
  • https://stanford.zoom.us/j/93701750367?pwd=VnlZeCt3NkVYUkRhK2VYSURtN2laQT09

BIODS 260
11/3/22
1:30 pm-2:50 pm
David Van Valen
MSOB x303

Title: Everything as Code

Bio: David Van Valen is an Assistant Professor in the Division of Biology and Bioengineering at Caltech. Before becoming faculty, he studied mathematics (B.S. 2003) and physics (B.S. 2003) at the Massachusetts Institute of Technology, applied physics (Ph.D. 2011) at Caltech, medicine (M.D. 2013) at UCLA, and bioengineering as a postdoctoral fellow at Stanford University. At Caltech, his research group develops new technologies at the intersection of imaging, genomics, and machine learning to produce quantitative measurements of living systems with single-cell resolution. David is the recipient of several awards, including a Hertz Fellowship (2005), a Rita Allen Scholar award (2020), A Pew-Stewart Cancer Research Scholar award (2021), a Heritage Medical Research Investigator award (2021), a Moore Inventor Fellowship (2021), and the NIH New Innovator award (2022).

Abstract: Biological systems are difficult to study because they consist of tens of thousands of parts, vary in space and time, and their fundamental unit—the cell—displays remarkable variation in its behavior. These challenges have spurred the development of genomics and imaging technologies over the past 30 years that have revolutionized our ability to capture information about biological systems in the form of images. Excitingly, these advances are poised to place the microscope back at the center of the modern biologist’s toolkit. Because we can now access temporal, spatial, and “parts list” variation via imaging, images have the potential to be a standard data type for biology.

For this vision to become reality, biology needs a new data infrastructure. Imaging methods are of little use if it is too difficult to convert the resulting data into quantitative, interpretable information. New deep learning methods are proving to be essential to reliable interpretation of imaging data. These methods differ from conventional algorithms in that they learn how to perform tasks from labeled data; they have demonstrated immense promise, but they are challenging to use in practice. The expansive training data required to power them are sorely lacking, as are easy-to-use software tools for creating and deploying new models. Solving these challenges through open software is a key goal of the Van Valen lab. In this talk, I describe DeepCell, a collection of software tools that meet the data, model, and deployment challenges associated with deep learning. These include tools for distributed labeling of biological imaging data, a collection of modern deep learning architectures tailored for biological image analysis tasks, and cloud-native software for making deep learning methods accessible to the broader life science community. I discuss how we have used DeepCell to label large-scale imaging datasets to power deep learning methods that achieve human level performance and enable new experimental designs for imaging-based experiments.

Website: https://vanvalen.caltech.edu

Zoom info:

PDF Flier