Published Apr 29, 2021

Polly Fordyce — Microfluidic Platforms and Machine Learning

Join Polly Fordyce as she delves into the revolutionary use of microfluidic platforms in understanding protein structures and their genetic variations, while exploring the groundbreaking potential of merging machine learning with biology to enhance scientific discoveries.
Episode Highlights
Gradient Dissent - A Machine Learning Podcast logo

Popular Clips

Episode Highlights

  • Microfluidic Devices

    introduces the concept of microfluidic devices, likening them to integrated circuits for fluid computations in biology. These devices revolutionize biological experiments by allowing high-throughput fluidic computations in a compact format, reducing the need for large, expensive equipment. Polly explains, "We can generate data at a scale that allows us to quantitatively test predictions from our colleagues in ML" 1. This innovation not only cuts costs but also enhances the interface between biology and machine learning, enabling more precise data collection and analysis 2.

       

    Enzyme Analysis

    Microfluidics play a crucial role in enzyme analysis, offering insights into enzyme activity and potential applications. describes how these devices allow for the creation of enzyme variants in tiny chambers, enabling researchers to assess mutations' effects on enzyme function. She notes, "This might help us classify variants in the human population for whether or not they're likely to compromise function and cause disease" 3. This technology not only aids in understanding enzyme behavior but also holds promise for environmental and medical applications 4.

       

    Fluidic Computation

    Fluidic computation, as explained by , offers a scalable and efficient alternative to traditional biological experiments. By using microfluidic devices, researchers can perform thousands of reactions simultaneously, significantly reducing time and resource consumption. Polly emphasizes, "These devices make it possible to use fewer reagents. So everything is low cost" 1. This approach not only accelerates data generation but also facilitates collaboration with machine learning experts to refine predictive models and explore new scientific frontiers 1.

Related Episodes