MicroBinfie Podcast, 117 From Math to Metagenomics - Titus Brown on Career Journeys and Software Solutions

Released on December 7, 2023

In this episode of the Micro Binfie Podcast, hosts Andrew Page and Lee Katz interview Titus Brown about his academic and professional journey. The discussion covers his transition from studying mathematics and physics as an undergraduate to becoming a bioinformatician, specializing in metagenomics and software development.

Topics Discussed

• Titus' Background:

  • Initial studies in math and physics.
  • Research in digital evolution and developmental biology.

• Transition to Bioinformatics:

  • Entered bioinformatics to address the increasing influx of genomic data in the 1990s.

• Tool Development:

  • Creation of early tools for comparative genomics and sequence analysis.
  • Emphasizing the philosophy of developing user-friendly software with comprehensive documentation.

• Research Contributions:

  • Contributions to transcriptomics, metagenomics, and k-mers at Michigan State.
  • Innovations in digital normalization and handling large sequencing datasets.

• Career at UC Davis:

  • Continued work on metagenomics and software development, specifically tools like khmer and sourmash.

• Challenges in Science:

  • Discussion on the challenges related to data reuse and accessibility.

Notable Publications

• Spacegraphcats:

  • Read the paper

• Sourmash:

  • Read the preprint

• IBD Exploration:

  • View the project

This episode offers valuable insights into the world of bioinformatics and the innovative approaches being taken to address challenges in genomic data analysis.

Extra notes

• The podcast discusses the evolution and application of k-mer analysis in bioinformatics, particularly for handling large-scale genomic data.
• Digital normalization is highlighted as a technique for managing data from deeply sequenced genomic datasets, which helps prevent computational workload overload by estimating coverage and discarding redundant reads.
• SourMash is a tool developed for MinHash-based analysis of k-mers, allowing efficient storage and comparison of genomic data by discarding 99.9% of k-mers, reducing memory and disk space requirements.
• The challenges of data reuse, interoperability, and socio-technical solutions in the context of FAIR (Findability, Accessibility, Interoperability, and Reusability) principles in bioinformatics are discussed.
• Collaboration and mentorship through open source platforms and tutorials play a crucial role in advancing the field, as evidenced by the uptake of developed tools without direct researcher involvement.
• The discussion briefly touches on a new method called Gambit from David Hess's lab, which focuses on targeted k-mer analysis, using specific prefixes to facilitate more accurate bacterial typing.
• The speaker mentions their involvement in large NIH consortia, emphasizing practical infrastructure implementation for large-scale data reuse, which underlines the complexity of global data sharing initiatives.
• The podcaster shares personal reflections on creativity and career evolution, acknowledging that career progression is often non-linear, driven by curiosity and adaptability rather than formal training alone.

Episode 117 transcript