Abstract:
This workshop caters to beginners and intermediates aiming to master cell segmentation in Optical Pooled Screens (OPS) by leveraging Python's flexibility and robust libraries like Stardist and Cellpose. Python's extensive library ecosystem and seamless integration with deep learning frameworks make it an ideal choice for efficient algorithm development.
Contrasting with R, Python offers a broader range of specialized image processing and deep learning libraries like scikit-image, OpenCV, TensorFlow, and PyTorch. These tools facilitate intuitive implementation of advanced segmentation techniques and deep learning models like U-Net, ensuring superior accuracy in cell segmentation tasks.
Participants will explore practical applications of Stardist and Cellpose in OPS image preprocessing and segmentation, emphasizing the advantages of utilizing Python over R for these purposes. The workshop aims to empower attendees with the ability to efficiently navigate cell segmentation challenges using Python, enabling them to achieve precise results and unlock the potential of OPS data for various biological analyses.
Instructor:
Jose Sergio Hleap Lozano
Passionate about unlocking biological insights through cutting-edge data analysis and leveraging big data within the bioinformatics domain, I specialize in the intersection of biology and data science. Equipped with a Ph.D. in Biochemistry and Molecular Biology from Dalhousie University, I excel at navigating High-Performance Computing (HPC) and cloud environments, using Python, Bash, and R.
My expertise extends to orchestrating infrastructure as a service (IaaS) solutions, optimizing computational workflows, and ensuring seamless operations for large-scale omics data analysis. An effective communicator, I bridge the gap between diverse stakeholders, including computer scientists, wet lab researchers, and business professionals, ensuring a cohesive and collaborative approach to problem-solving. Throughout my journey, I've contributed to prestigious jornals such as Nature and Proc. R. Soc. B, demonstrating my commitment to advancing genomic understanding through data-driven methodologies.
