Niels Bantilan

Unlocking value from sequencing data with workflow management

Unlocking value from sequencing data with workflow management

Sequencing data is the foundation of many biotech companies today as it serves as a blueprint at the molecular level — from RNA, DNA, to proteins. Advances in sequencing and computing have made this data more accessible for various applications like drug discovery, precision medicine, synthetic biology, and agriculture. However, analyzing such massive sequencing files is challenging and requires significant management systems. Workflow management is commonly used to ensure data integrity and promote collaboration between scientists and engineers.

In this panel discussion, hear from experts in the biotech industry discuss their unique challenges, workflow strategies, and future trends in sequencing technologies and workflow management.

Discussion points

  • Data Characteristics: Discover the various data types, formats, file sizes, and the sheer scale of data that they deal with on a daily basis.
  • Tools and frameworks: Learn about the specific libraries, frameworks, and tools used to extract meaningful insights from genomic data.
  • Data Infrastructure Challenges: Explore the hurdles faced in setting up and maintaining data infrastructure and how they tackle these challenges head-on.
  • Translating Raw Data to Insights: Understand the biggest challenges in the journey from raw sequencing data to valuable insights, and the strategies employed to overcome them.
  • Role of Workflow Management: Learn how workflow management plays a pivotal role in streamlining their processes and optimizing bioinformatics and machine learning pipelines.
  • Unique and Shared Requirements: Discover the unique and shared requirements of their pipelines, and how they tailor their workflows to meet these demands.

Meet the panelists

  • Jeev Balakrishnan, Software Engineer at Union: Biologist turned software engineer with a decade of experience building and operating research and production platforms at scale, across companies moving the needle in the life sciences and healthcare. Currently at, empowering innovators with the right primitives to build with minimal friction.
  • Eli Bixby, Co-founder and Machine Learning Engineer at Cradle: Eli is a co-founder at Cradle, where he makes sure their models and algorithms are doing what they are designed to do, and he keeps an eye out for the latest and greatest techniques in the literature. He was previously at Google (Brain, Accelerated Science, Cloud) working on biological sequence design, AutoML, and natural language understanding. He studied mathematics, computer science, and biochemistry.
  • Alex Ford, Head of Data Platform at AbCellera: Alex is the Head of Data Architecture at AbCellera Biologics, a full-stack antibody discovery and development firm in Vancouver BC. His team builds a data science platform integrating laboratory automation, bioinformatics, computational protein engineering and data analytics to support bringing antibody-based medicines from target to clinic.
  • Brian O’Donovan, Head of Bioinformatics at Delve Bio: Bay Area native. Undergrad in BioEngineering from UCSD.  Worked on pacemaker/defibrillator diagnostic and therapeutic/interventional algorithms before returning to grad school for a PhD in Bioinformatics at UCSF. PhD Thesis was on the clinical applications of genomic sequencing technologies/assays — specifically aimed at tough-to-diagnose cases of meningitis/encephalitis. Worked in biotech since graduating: Audentes Therapeutics (now Astellas) gene therapy program (2 years), Freenome for early colorectal cancer detection (4 years), and now at Delve - a company spun out of work that germinated in my old lab at UCSF. We are working to commercialize and scale a clinical metagenomics test pioneered by our co-founders at UCSF.
  • Thomas Vetterli, Director of Machine Learning and Bioinformatics at Hedera Dx: Thomas is a dynamic engineer specializing in machine learning and bioinformatics platforms for healthcare and life sciences. Currently serving as the Director of Machine Learning and Bioinformatics at HederaDx, he leads a team in developing NGS tumor monitoring pipelines for liquid biopsies. With previous roles at Freenome and Berkeley Lights, Thomas participated in the development of machine learning platforms for early cancer detection and single cell automation.