The software layer for biotech | Latch Bio

Episode 17 of S³

Very soon, biological data will fly by all 120 zettabytes of the internet's data. What software will biology use for all of it?

Enter the feature startup for episode 17 of S³, Latch Bio.

Say hello to S³’s first sponsor: Pillar VC

I take S³ sponsorship super seriously, we don’t agree to sponsorship lightly and that’s why I’m super excited to have such an amazing partner in Pillar VC as our first ever sponsorship.

Pillar is single handedly increasing the number of biotech companies that exist in the world in a meaningful way. They’re pioneering a “Founder-led bio” approach to biotech and investing in teams that are founder led rather than brought about by large institutions. Check out Pillar’s website here.

Watch Episode 17! Then read this to go deep.

Once you watch the episode you can expect this blog post to go deeper on:

  • How modern day bio is done & the lack of software for it

  • What Latch unlocks today…

  • The future features Latch is building for tomorrow

New age bio needs modern software

The sheer volume of data generated by modern biotechnology is nothing short of staggering. Next-generation sequencing and multiplexing have catapulted the field into a new era, where the data generated is not just voluminous but deeply complex. Traditional biotech research methods, which once relied on slower, more manual processes, pale in comparison to the torrents of data unleashed by these new technologies. This isn't just a quantitative change; it's a qualitative one, fundamentally altering the nature of biotech research.

This data explosion presents a unique challenge: traditional software and data management tools are woefully inadequate for this new landscape. The complexity and scale of the data generated by next-gen sequencing and multiplexing require not just larger storage capacities, but entirely new paradigms for processing, analyzing, and making sense of this information. Existing tools are like trying to navigate a modern metropolis with a map designed for a medieval village – they simply can't capture the complexity and nuance of this new world.

Enter Latch.

Latch has been designed from the ground up for the scale and complexity of modern biotech data, designed not just to manage but to harness this data. It's a solution that finally matches the scale of the problem, a tool that turns what was once an overwhelming challenge into an unprecedented opportunity for discovery and innovation.

So far in S³ history the Latch boys have won for the prize the most entertaining and hilarious cofounder photos.

Latch was founded by Alfredo Andere (CEO), Kyle Giffin (COO), and Kenny Workman (CTO), who met as undergraduates at the University of California, Berkeley. Alfredo worked as a Data Engineer at Facebook and as a Software Engineer at Google Brain prior to founding Latch. Kyle Giffin was a neuroscience researcher and Kenny was working in bio wet labs at the age of 15... I didn’t ask while filming how the hell that happened.

Despite their youth, these guys are serious. When setting up to interview, Kenny shared that “it’s been 3 years of constant executing,” you could tell from his tone he was serious, you can’t really fake that kind of exhausted yet fiercely dedicated tone if you haven’t been in the trenches.

What Latch unlocks

Organization for unlimited data storage

There’s a great meme on Twitter about how hard it is to put together a functioning Google Drive organization system for one’s projects, now how about doing file organization for zettabytes of data?

Latch’s file system allows the storage of unlimited amounts of raw data, breaking down the barriers of data silos. This system enhances the accessibility of cloud data, analysis results, and instruments, making them queryable for all team members — that’s a big deal in bio, normally you have to talk to specific team members to access specific data. The simplicity of the interface enables researchers themselves to explore results and make faster decisions on their experiments.

Latch's database feature is its cornerstone for organizing metadata efficiently. It provides a platform where experiments can be explored through interactive tables, this feature not only helps in visualizing and understanding an overview of R&D progress but also offers scientists an intuitive interface for filling out metadata with custom quality control. The ability to programmatically query these tables during workflow runs signifies a leap in how metadata is organized and utilized in research environments

Software customizability without the engineering

The Workflow Manager in Latch is a game-changer in deploying and executing custom workflows. It allows users to set up these workflows in minutes and scale them to thousands of cores, providing unprecedented control over analysis. This feature represents a significant cost-saving measure by automatically running workflows without programming knowledge, empowering researchers to take control of their analyses through a simple user interface.

As Kyle mentions in the episode, this also saves significant software engineering headcount and lets small biotechs onto a more even playing field with larger biotechs.

Elastic compute

Elastic compute is one of the most striking features Latch offers. It sets up flexible compute instances with pre-configured analysis notebooks, providing access to supercomputer-level power on-demand. This feature allows users to switch on and off the compute resources easily or set an auto-shut off to eliminate waste. By leveraging pre-configured Jupyter Notebooks and RStudio, it enables researchers to focus on the actual analysis, thus exemplifying how technology can simplify and amplify the power of scientific computation

Future Latch features

Latch is venturing into uncharted territories of data management, how many companies have to deal with zettabytes of data? Not many...

As a result, the team are thinking how compute should be approached for data of unprecedented scale, a task as daunting as it is crucial. Latch's ambition isn't just to manage this “data deluge” — it's to transform it into a wellspring of accessible, actionable knowledge. This is a classic example of turning a formidable challenge into an opportunity for groundbreaking innovation.

A particularly exciting avenue Latch is exploring is making publicly bio available data queryable. Consider the vast amounts of data from previous experiments, lying dormant, waiting to be leveraged. Latch aims to synthesize this data and make it queryable in their platform. The impact? A seismic shift in how biologists and researchers access, interpret, and utilize existing data. This isn't just about building a better tool; it's about fundamentally changing the landscape of biotech research.

Latch's trajectory is beyond exciting: identifying a deep, intricate problem and applying problem-solving creativity to it over time. I believe what Latch is building will redefine today’s, and perhaps even more importantly, tomorrow’s biotech industry.

Keep on building the future,

— Jason

Filmed in San Francisco, CA | Edited in San Francisco, CA | Composed in Davis, CA